Artificial Intelligence (Part-VIII)- Deep Learning in Artificial Intelligence

Image for representative purpose only.

Know in Details About Deep Learning in Artificial Intelligence

Here we continue with the last part of our blog on artificial intelligence. Those who have missed our seventh part can read it from Here. It will help to connect with this last part of the blog discussing about deep learning in artificial intelligence. Let us explore the blog to find out in more details. In word of Steve Polyak:

"Before we work on artificial intelligence why don’t we do something about natural stupidity? So, how would you weigh in? What’s your opinion about artificial intelligence?"

What is Deep Learning?

Deep learning is any artificial neural network that can learn a long chain of causal links. For example, a feedforward network with six hidden layers can learn a seven-link causal chain (six hidden layers + output layer) and has a "credit assignment path" (CAP) depth of seven. Many deep learning systems need to be able to learn chains ten or more causal links in length. Deep learning has transformed many important subfields of artificial intelligence, including computer vision, speech recognition, natural language processing and others. According to one overview,  the expression "Deep Learning" was introduced to the Machine Learning community by Rina Dechter in 1986 and gained traction after Igor Aizenberg and colleagues introduced it to Artificial Neural Networks in 2000. The first functional Deep Learning networks were published by Alexey Grigorevich Ivakhnenko and V. G. Lapa in 1965. These networks are trained one layer at a time. Ivakhnenko's 1971 paper describes the learning of a deep feedforward multilayer perceptron with eight layers, already much deeper than many later networks. In 2006, a publication by Geoffrey Hinton and Ruslan Salakhutdinov introduced another way of pre-training many-layered feedforward neural networks (FNNs) one layer at a time, treating each layer in turn as an unsupervised restricted Boltzmann machine, then using supervised back propagation for fine-tuning. Similar to shallow artificial neural networks, deep neural networks can model complex non-linear relationships. Over the last few years, advances in both machine learning algorithms and computer hardware have led to more efficient methods for training deep neural networks that contain many layers of non-linear hidden units and a very large output layer. Deep learning often uses convolutional neural networks (CNNs), whose origins can be traced back to the Neocognitron introduced by Kunihiko Fukushima in 1980. In 1989, Yann LeCun and colleagues applied backpropagation to such an architecture. In the early 2000s, in an industrial application CNNs already processed an estimated 10% to 20% of all the checks written in the US. Since 2011, fast implementations of CNNs on GPUs have won many visual pattern recognition competitions. CNNs with 12 convolutional layers were used in conjunction with reinforcement learning by Deepmind's "AlphaGo Lee", the program that beat a top Go champion in 2016.

Recurrent Neural Network

Early on, deep learning was also applied to sequence learning with recurrent neural networks (RNNs) which are in theory Turing complete and can run arbitrary programs to process arbitrary sequences of inputs. The depth of an RNN is unlimited and depends on the length of its input sequence; thus, an RNN is an example of deep learning. RNNs can be trained by gradient descent but suffer from the vanishing gradient problem. In 1992, it was shown that unsupervised pre-training of a stack of recurrent neural networks can speed up subsequent supervised learning of deep sequential problems. Numerous researchers now use variants of a deep learning recurrent NN called the long short-term memory (LSTM) network published by Hochreiter & Schmidhuber in 1997. LSTM is often trained by Connectionist Temporal Classification (CTC). At Google, Microsoft and Baidu this approach has revolutionised speech recognition. For example, in 2015, Google's speech recognition experienced a dramatic performance jump of 49% through CTC-trained LSTM, which is now available through Google Voice to billions of smartphone users. Google also used LSTM to improve machine translation, Language Modelling and Multilingual Language Processing. LSTM combined with CNNs also improved automatic image captioning and a plethora of other applications.

Artificial Intelligence (Part-VII)- Machine Learning

Image for representative purpose only.

Statistical Classification of Artificial Intelligence- Machine Learning

Here we continue with the seventh part of our blog on artificial intelligence. Those who have missed our sixth part can read it from Here. It will help to connect with this seventh part of the blog discussing about statistical classification of artificial intelligence popularly known as Machine Learning. Let us explore the blog to find out about it in more details. In words of James Barrat:

"I don’t want to really scare you, but it was alarming how many people I talked to who are highly placed people in AI who have retreats that are sort of 'bug out' houses, to which they could flee if it all hits the fan".

The simplest AI applications can be divided into two types: classifiers ("if shiny then diamond") and controllers ("if shiny then pick up"). Controllers do, however, also classify conditions before inferring actions, and therefore classification forms a central part of many AI systems. Classifiers are functions that use pattern matching to determine a closest match. They can be tuned according to examples, making them very attractive for use in AI. These examples are known as observations or patterns. In supervised learning, each pattern belongs to a certain predefined class. A class can be seen as a decision that has to be made. All the observations combined with their class labels are known as a data set. When a new observation is received, that observation is classified based on previous experience. A classifier can be trained in various ways; there are many statistical and machine learning approaches. The decision tree is perhaps the most widely used machine learning algorithm. Other widely used classifiers are the neural network, k-nearest neighbour algorithm, kernel methods such as the support vector machine (SVM), Gaussian mixture model, and the extremely popular naive Bayes classifier. Classifier performance depends greatly on the characteristics of the data to be classified, such as the dataset size, the dimensionality, and the level of noise. Model-based classifiers perform well if the assumed model is an extremely good fit for the actual data. Otherwise, if no matching model is available, and if accuracy (rather than speed or scalability) is the sole concern, conventional wisdom is that discriminative classifiers (especially SVM) tend to be more accurate than model-based classifiers such as "naive Bayes" on most practical data sets.

What is Artificial Neutral Network?

Neural networks, or neural nets, were inspired by the architecture of neurons in the human brain. A simple "neuron" N accepts input from multiple other neurons, each of which, when activated (or "fired"), cast a weighted "vote" for or against whether neuron N should itself activate. Learning requires an algorithm to adjust these weights based on the training data; one simple algorithm (dubbed "fire together, wire together") is to increase the weight between two connected neurons when the activation of one triggers the successful activation of another. The net forms "concepts" that are distributed among a subnetwork of shared neurons that tend to fire together; a concept meaning "leg" might be coupled with a subnetwork meaning "foot" that includes the sound for "foot". Neurons have a continuous spectrum of activation; in addition, neurons can process inputs in a nonlinear way rather than weighing straightforward votes. Modern neural nets can learn both continuous functions and, surprisingly, digital logical operations. Neural networks' early successes included predicting the stock market and (in 1995) a mostly self-driving car. In the 2010s, advances in neural networks using deep learning thrust AI into widespread public consciousness and contributed to an enormous upshift in corporate AI spending; for example, AI-related M&A in 2017 was over 25 times as large as in 2015.

The study of non-learning artificial neural networks began in the decade before the field of AI research was founded, in the work of Walter Pitts and Warren McCullouch. Frank Rosenblatt invented the perceptron, a learning network with a single layer, similar to the old concept of linear regression. Early pioneers also include Alexey Grigorevich Ivakhnenko, Teuvo Kohonen, Stephen Grossberg, Kunihiko Fukushima, Christoph von der Malsburg, David Willshaw, Shun-Ichi Amari, Bernard Widrow, John Hopfield, Eduardo R. Caianiello, and others. The main categories of networks are acyclic or feedforward neural networks (where the signal passes in only one direction) and recurrent neural networks (which allow feedback and short-term memories of previous input events). Among the most popular feedforward networks are perceptrons, multi-layer perceptrons and radial basis networks. Neural networks can be applied to the problem of intelligent control (for robotics) or learning, using such techniques as Hebbian learning , GMDH or competitive learning.

Continued in the next blog...

Artificial Intelligence (Part-VI)- Amazing Tools Used in Artificial Intelligence

Image for representative purpose only.

Find Out Some Of the Amazing Tools Used For the Development of Artificial Intelligence

Now we continue with the sixth part of our blog on artificial intelligence. Those who have missed our fifth part can read it from Here. It will help to connect with this sixth part of the blog discussing some of the important tools and techniques used in the development of artificial intelligence. Let us explore the blog to find out in more details. In words of Larry Page:

"Artificial intelligence would be the ultimate version of Google. The ultimate search engine that would understand everything on the web. It would understand exactly what you wanted, and it would give you the right thing. We're nowhere near doing that now. However, we can get incrementally closer to that, and that is basically what we work on".

Tools of Artificial Intelligence

Some more tools used in the development of artificial intelligence with higher accuracy and better performance are-

ii) Mathematical Optimization or Search Optimization

Many problems in AI can be solved in theory by intelligently searching through many possible solutions: Reasoning can be reduced to performing a search. For example, logical proof can be viewed as searching for a path that leads from premises to conclusions, where each step is the application of an inference rule. Planning algorithms search through trees of goals and sub-goals, attempting to find a path to a target goal, a process called means-ends analysis. Robotics algorithms for moving limbs and grasping objects use local searches in configuration space. Many learning algorithms use search algorithms based on optimization. Simple exhaustive searches are rarely sufficient for most real-world problems: the search space (the number of places to search) quickly grows to astronomical numbers. The result is a search that is too slow or never completes. The solution, for many problems, is to use "heuristics" or "rules of thumb" that prioritize choices in favour of those that are more likely to reach a goal and to do so in a shorter number of steps. In some search methodologies heuristics can also serve to entirely eliminate some choices that are unlikely to lead to a goal (called "pruning the search tree"). Heuristics supply the program with a "best guess" for the path on which the solution lies. Heuristics limit the search for solutions into a smaller sample size. 

A very different kind of search came to prominence in the 1990s, based on the mathematical theory of optimization. For many problems, it is possible to begin the search with some form of a guess and then refine the guess incrementally until no more refinements can be made. These algorithms can be visualized as blind hill climbing: we begin the search at a random point on the landscape, and then, by jumps or steps, we keep moving our guess uphill, until we reach the top. Other optimization algorithms are simulated annealing, beam search and random optimization. Evolutionary computation uses a form of optimization search. For example, they may begin with a population of organisms (the guesses) and then allow them to mutate and recombine, selecting only the fittest to survive each generation (refining the guesses). Classic evolutionary algorithms include genetic algorithms, gene expression programming, and genetic programming. Alternatively, distributed search processes can coordinate via swarm intelligence algorithms. Two popular swarm algorithms used in search are particle swarm optimization and this is inspired by bird flocking and ant colony optimization which is inspired by ant trails.

iii) Decision Theory and Utility Theory

Expectation-maximization clustering of Old Faithful eruption data starts from a random guess but then successfully converges on an accurate clustering of the two physically distinct modes of eruption. Many problems in AI (in reasoning, planning, learning, perception, and robotics) require the agent to operate with incomplete or uncertain information. AI researchers have devised a number of powerful tools to solve these problems using methods from probability theory and economics. Bayesian networks are a very general tool that can be used for a large number of problems: reasoning (using the Bayesian inference algorithm),  learning (using the expectation-maximization algorithm), planning (using decision networks) and perception (using dynamic Bayesian networks). Probabilistic algorithms can also be used for filtering, prediction, smoothing and finding explanations for streams of data, helping perception systems to analyse processes that occur over time. Compared with symbolic logic, formal Bayesian inference is computationally expensive. For inference to be tractable, most observations must be conditionally independent of one another. Complicated graphs with diamonds or other "loops" (undirected cycles) can require a sophisticated method such as Markov Chain Monte Carlo, which spreads an ensemble of random walkers throughout the Bayesian network and attempts to converge to an assessment of the conditional probabilities. Bayesian networks are used on XBox Live to rate and match players; wins and losses are "evidence" of how good a player is. A key concept from the science of economics is "utility": a measure of how valuable something is to an intelligent agent. Precise mathematical tools have been developed that analyse how an agent can make choices and plan, using decision theory, decision analysis, and information value theory. These tools include models such as Markov decision processes, dynamic decision networks, game theory and mechanism design.

Continued in the next blog...

Artificial Intelligence (Part-V)- Integrating the Approaches and Tools of AI

Image for representative purpose only.

The Different Approaches and Tools Used in Artificial Intelligence

Here we continue with the fifth part of our blog on artificial intelligence. Those who have missed our fourth part can read it from Here. It will help to connect with this fifth part of the blog discussing on the different approaches and tools used in the development of artificial intelligence. Let us explore the blog to read about these in more details. In words of Elon Musk: 

"The pace of progress in artificial intelligence (I’m not referring to narrow AI) is incredibly fast. Unless you have direct exposure to groups like Deepmind, you have no idea how fast—it is growing at a pace close to exponential. The risk of something seriously dangerous happening is in the five-year timeframe. 10 years at most".  

Intelligent Agent Paradigm

An intelligent agent is a system that perceives its environment and takes actions which maximize its chances of success. The simplest intelligent agents are programs that solve specific problems. More complicated agents include human beings and organizations of human beings (such as firms). The paradigm allows researchers to directly compare or even combine different approaches to isolated problems, by asking which agent is best at maximizing a given "goal function". An agent that solves a specific problem can use any approach that works – some agents are symbolic and logical, some are sub-symbolic artificial neural networks and others may use new approaches. The paradigm also gives researchers a common language to communicate with other fields—such as decision theory and economics—that also use concepts of abstract agents. Building a complete agent requires researchers to address realistic problems of integration; for example, because sensory systems give uncertain information about the environment, planning systems must be able to function in the presence of uncertainty. The intelligent agent paradigm became widely accepted during the 1990s.

Agent Architectures and Cognitive Architectures

Researchers have designed systems to build intelligent systems out of interacting intelligent agents in a multi-agent system. A hierarchical control system provides a bridge between sub-symbolic AI at its lowest, reactive levels and traditional symbolic AI at its highest levels, where relaxed time constraints permit planning and world modelling. Some cognitive architectures are custom-built to solve a arrow problem; others, such as Soar, are designed to mimic human cognition and to provide insight into general intelligence. Modern extensions of Soar are hybrid intelligent systems that include both symbolic and sub-symbolic components.

Tools of Artificial Intelligence:

Some of the different tools used in the development of artificial intelligence with higher accuracy and better performance are-

i) Logic Programming and Automated Reasoning

Logic is used for knowledge representation and problem solving, but it can be applied to other problems as well. For example, the satplan algorithm uses logic for planning and inductive logic programming is a method for learning. Several different forms of logic are used in AI research. Propositional logic involves truth functions such as "or" and "not". First-order logic adds quantifiers and predicates, and can express facts about objects, their properties, and their relations with each other. Fuzzy set theory assigns a "degree of truth" (between 0 and 1) to vague statements such as "Rohit is old" (or rich, or tall, or hungry) that are too linguistically imprecise to be completely true or false. Fuzzy logic is successfully used in control systems to allow experts to contribute vague rules such as "if you are close to the destination station and moving fast, increase the train's brake pressure"; these vague rules can then be numerically refined within the system. Fuzzy logic fails to scale well in knowledge bases; many AI researchers question the validity of chaining fuzzy-logic inferences. Default logic, non-monotonic logic and circumscription are forms of logic designed to help with default reasoning and the qualification problem. Several extensions of logic have been designed to handle specific domains of knowledge, such as: description logic; situation calculus, event calculus and fluent calculus (for representing events and time); causal calculus; belief calculus; and modal logic. Overall, qualitative symbolic logic is brittle and scales poorly in the presence of noise or other uncertainty. Exceptions to rules are numerous, and it is difficult for logical systems to function in the presence of contradictory rules or regulations.

Continued in the next blog...

Artificial Intelligence (Part-IV)- Arms of AI and Soft Computing

Image for representative purpose only.

What is Embodied Intelligence? How this coincide with the development of the Embodiment Thesis in the related field of Cognitive Science? What is Computational Intelligence and Soft Computing? How is Statistical Learning done?

Now we continue with the fourth part of our blog on artificial intelligence. Those who have missed our third part can read it from Here. It will help to connect with this fourth part of the blog discussing about embodied intelligence, its coincidence with the development of embodiment thesis in the field of cognitive science, computational intelligence, soft computing and statistical learning. Let us explore the blog to read about these in more details. In words of Claude Shannon:

"I visualize a time when we will be to robots what dogs are to humans, and I’m rooting for the machines".

Embodied Intelligence

This includes embodied, situated, behaviour-based, and nouvelle AI. Researchers from the related field of robotics, such as Rodney Brooks, rejected symbolic AI and focused on the basic engineering problems that would allow robots to move and survive. Their work revived the non-symbolic viewpoint of the early cybernetics researchers of the 1950s and reintroduced the use of control theory in AI. This coincided with the development of the embodied mind thesis in the related field of cognitive science: the idea that aspects of the body are required for higher intelligence.

Within developmental robotics, developmental learning approaches are elaborated upon to allow robots to accumulate repertoires of novel skills through autonomous self-exploration, social interaction with human teachers, and the use of guidance mechanisms.

Computational Intelligence and Soft Computing

Ideas or concept in neural networks and "connectionism" was revived by David Rumel hart and others in the middle of the 1980s.[157] Artificial neural networks are an example of soft computing --- they are solutions to problems which cannot be solved with complete logical certainty, and where an approximate solution is often sufficient. Other soft computing approaches to AI include fuzzy systems, evolutionary computation and many statistical tools. The application of soft computing to AI is studied collectively by the emerging discipline of computational intelligence.

Statistical Learning

Much of traditional GOFAI got bogged down on ad hoc patches to symbolic computation that worked on their own toy models but failed to generalize to real-world results. However, around the 1990s, AI researchers adopted sophisticated mathematical tools, such as hidden Markov models (HMM), information theory, and normative Bayesian decision theory to compare or to unify competing architectures. The shared mathematical language permitted a high level of collaboration with more established fields (like mathematics, economics or operations research). Compared with GOFAI, new "statistical learning" techniques such as HMM and neural networks were gaining higher levels of accuracy in many practical domains such as data mining, without necessarily acquiring semantic understanding of the datasets. The increased successes with real-world data led to increasing emphasis on comparing different approaches against shared test data to see which approach performed best in a broader context than that provided by idiosyncratic toy models; AI research was becoming more scientific. Nowadays results of experiments are often rigorously measurable, and are sometimes reproducible. Different statistical learning techniques have different limitations; for example, basic HMM cannot model the infinite possible combinations of natural language. Critics note that the shift from GOFAI to statistical learning is often also a shift away from Explainable AI. In AGI research, some scholars caution against over-reliance on statistical learning, and argue that continuing research into GOFAI will still be necessary to attain general intelligence.

Continued in the next blog...

Artificial Intelligence (Part-III)- Developing Artificial Intelligence With Systematic Planning and Learning

Image for representative purpose only.

Systematic Knowledgebase with Common Sense are the Corner-Stone of Artificial Intelligence

Today we continue with the third part of our blog on artificial intelligence. Those who have missed our second blog can read it from Here. It will help to connect with this third part of the blog discussing the importance of proper and systematic use of knowledgebase and common sense to develop artificial intelligence.

Knowledge Engineering: The Corner-Stone to Classical Artificial Research

There are some expert systems to collect together accurate knowledge possessed by experts in some narrow domain. In addition, some projects attempt to collect the "fundamental knowledge" known to the average person into a database containing comprehensive knowledge about the world. Among the things a comprehensive fundamental knowledge base would contain are: objects, properties, categories and relations between objects; situations, events, states and time;  causes and effects; knowledge about knowledge and many other, less well researched domains. A depiction of "what exists" is an ontology: the set of objects, relations, approach, and properties formally described so that software agents can interpret them. The semantics of these are captured as description logic concepts, roles, and individuals, and typically implemented as classes, properties, and individuals in the Web Ontology Language. The most general ontologies are called upper ontologies, which attempt to provide a foundation for all other knowledge by acting as mediators between domain ontologies that cover specific knowledge about a particular knowledge domain (field of interest or area of concern). Such formal knowledge representations can be used in content-based indexing and retrieval, scene analysis, clinical judgement support, knowledge discovery,  and other areas.

Artificial Intelligence: Broad Combination of Common Sense and Knowledgebase

It’s within the grip of common people is to represent the knowledge as "facts" or "statements" that they could express verbally. For example, a chess master will avoid a particular chess position because it "feels too exposed" or an art authority can take one look at a statue and realize that it is a fictitious. These are non-conscious and sub-symbolic intuitions or tendencies in the human brain. Knowledge like this apprise, guide and provides a context for symbolic, cognizant knowledge. As with the related problem of sub-symbolic reasoning, it is hoped that situated AI, computational intelligence, or statistical AI will provide ways to perform this type of knowledge. The count of atomic facts that the moderate person knows is very huge. Research projects that pursuit to build a complete knowledge base of common-sense knowledge require huge amounts of laborious ontological engineering they must be develop, manually, one complicated concept at a time.

How Planning and Learning is done in Artificial Intelligence?

Multi-agent planning uses the assistance and counteraction of many agents to fulfil a given target. Appearing behaviour such as this is used by evolutionary algorithms and swarm intelligence. In humanistic planning problems, the agent can speculate that it is the only system acting in the world, allowing the agent to be certain of the countercoup of its actions. However, if the agent is not the only actor, then it requires that the agent can acumen under uncertainty. This calls for an agent that can not only assess its environment and make predictions, but also appraise its predictions and adapt based on its assessment. Computational learning theory can permit beginner by computational complexity, by sample complexity about the exact amount of data that is required, or by other notions of optimization. In brace learning the agent is rewarded for good feedback and punished for bad ones. The agent uses this sequence of rewards and punishments to form a strategy for running in its problem space. Unsupervised learning is the capability to search patterns in a torrent of input. Supervised learning includes both classification and numerical regression. Classification is used to determine what category something belongs in, after seeing a number of examples of things from several categories. Regression is the attempt to produce a function that describes the relationship between inputs and outputs and predicts how the outputs should change as the inputs change. Both classifiers and regression learners can be viewed as "function approximators" trying to learn an unknown (possibly implicit) function; for example, a spam classifier can be displayed as learning a function that maps from the written text of an email to one of two categories, "spam" or "not spam".

Intelligent agents must be capable to set targets and achieve them. They need a way to visualize the future a depiction of the state of the world and be able to make predictions about how their actions will change it and be able to make choices that maximize the utility (or "value") of available choices.

An Important Note to Always Keep in Mind 

Artificial Intelligence and the automated technology are one side of the life that always interest and wander us with the new ideas, topics, innovations, products …etc. AI is still not enforce as the films representing it(i.e. intelligent robots), however there are many important tries to reach the level and to challenge in market, like sometimes the robots that they show in TV. Nevertheless, the hidden projects and the advancements in industrial companies. 

Continued in the next blog...

Artificial Intelligence (Part-II)- Artificial Intelligence or Automated Machines?

Image for representative purpose only.

Artificial Intelligence or Automated Machines? Blessing or Curse to Our Mankind?

Now we continue with the second part of our blog on artificial intelligence. Those who have missed our first blog can read it from Here. It will help to connect with this second part of the blog discussing artificial intelligence or automated machines is what we are developing and is it a blessing or curse to mankind.

Artificial Intelligence can be a Good Problem Solver

In early days, researchers developed algorithms that burlesqued step-by-step reasoning that humans required when they solve puzzles or make logical deductions. In the late 1980s and 1990s AI research had built procedures for business with uncertain or incomplete data, employing concepts from probability and economics. It times of crisis, these algorithms proved wrong, since they are insufficient for solving larger critical and complicated problems because they experienced a combinatoric explosion. They became exponentially slower as the problems grew larger.  In fact, even humans barely use the step-by-step contemplation that early AI research was able to model. They solve most of their problems using fast, perceptive judgments.

Artificial Intelligence may act as a Social Intelligence

During the long terms and greater execution of process, social skills and an analyzing the  human emotion and game theory would be corners stone to a social agent. Being efficient to predict the actions of others by understanding their motives and emotional states would permit an agent to make better decisions. Some computer systems caricaturist human emotion and expressions to appear more sensitive to the emotional dynamics of human interaction, or to otherwise expedite human–computer interaction. Similarly, some virtual assistants are developed for doing conversations or even to chitchat amusingly; this tends to give ignorant users an impractical  or impossible conception of how knowledgeable existing computer agents actually are.  Medium successes related to effective computing include written text sentiment analysis and, more recently, multi-modal affect analysis (see multi-modal sentiment analysis), wherein AI classifies the affects advertised by a videotaped subject.

Building Motions in Artificially Intelligent Robots

In the year 1988,  Moravec's paradox concluded that low-level sensory motor dexterity that humans take for granted are, counter intuitively, difficult to build and programmed into a robot; the paradox is named after Hans Moravec. It is easy to make computers exhibit adult level performing systems on intelligence tests or playing checkers, and difficult or impossible to give them the skills of a one-year-old when it comes to perception and mobility. This is the parameters to the fact that, unlike checkers, physical dexterity has been a direct target of natural selection for millions of years. Motion planning is the method of breaking down a movement task into "anthropophagite" such as individual joint movements. Such movement often affects compliant motion, a method where movement requires maintaining physical contact with an object. Modern robotic arms and other industrial robots, widely used in modern  business, can learn from involvement how to move efficiently despite the presence of friction and gear slippage. A modern mobile robot, when given a small, static, and visible environment, can easily determine its location and map its environment; however, dynamic environments, such as (in endoscopy) the interior of a patient's breathing body, pose a greater challenge.

Artificial Intelligence can be Threatening

As per researchers they agreed that a super imaginative AI is unlikely to illustrate human emotions like loving someone hate someone or getting angry with someone, and that there is no reason to expect AI to become intentionally philanthropic or malignant. As per experts AI may become dangerous and there are two scenarios:

1. The AI is Programmed to do something Disastrous

Automatically operated tools are artificial intelligence systems that are developed to kill. In the hands of the terrorists, these weapons could easily cause mass destruction or even they can destruct a part of the nation. Moreover, an AI arms race could recklessly cause to an AI war that also outputs to destruction of human race. To avoid being circumvent by the enemy, these tools would be developed to be extremely difficult to simply “turn off,” so humans could apparently lose control of such a footing. This threat is one that’s present even with narrow Automated Systems or AI, but grows as levels of AI intelligence and self-determination increase.

2. The Artificial Intelligence is developed to do something Fruitful, but it develops a Destructive Method for achieving our target

This is strikingly difficult for us to achieve and it would be too much dangerous when we rely on AI to fulfill this type of goals. If we operate an obedient intelligent car to take you to the airport as early as possible, it might get you there drive away by helicopters and covered in regurgitate, doing not what you wanted but literally what you asked for. If a super brained and automated system is encumbered with an aggressive or determined geoengineering project, it might bring out catastrophe with our ecosystem as a side effect, and view human attempts to block it as a risk to be met.

Continued in the next blog...

Artificial Intelligence (Part-I)- The Driving Force of Modern Digitalization

Image for representative purpose only.

Introduce Yourself to the World of Artificial Intelligence, a Giant Leap Towards Modern Digitalization

Artificial Intelligence (AI) is the most popular and desirable tool for ground breaking development in almost every field of science and technology. This tool is continuously paving the way for modern digitalization thus providing better performance and greatly reducing human effort. We will present you with a series of blogs to know and explore about this interesting field.

Introducing Artificial Intelligence

Artificial intelligence is a kind of intelligence exhibit by machines in comparison to the natural intelligence (NI) spread out by humans and other animals. In computer science AI research is described as the study of "intelligent agents": any device that recognize its environment and takes actions that maximize its chance of successfully achieving its goals. Colloquially, the term "artificial intelligence" is enforced when a machine mimics "cognitive" functions that humans associate with other human minds, such as "learning" and "problem solving". Once the President of Future of Life Institute, Mr. Max Tegmark  quoted that-

"Everything we love about civilization is a product of intelligence, so amplifying our human intelligence with artificial intelligence has the potential of helping civilization flourish like never before – as long as we manage to keep the technology beneficial".

Starting from the world best AI driven applications up to self-driven car, AI is progressing rapidly. Artificial intelligence today is properly known as narrow AI (or weak AI), in that it is designed to perform a narrow task (e.g. only facial recognition or only internet searches or only driving a car). However, the long-term targets of many researchers is to develop general AI. While narrow AI may perform whatever its specific task is, like solving equations, AI would outperform humans at nearly every mental task. The ambit of AI is disputed: as machines become increasingly proficient, tasks considered as requiring "intelligence" are often eliminated from the definition, a method known as the AI effect, initializing to the quip, "AI is whatever hasn't been done yet." For instance, optical character recognition is frequently excluded from "artificial intelligence", having become a routine technology. Efficiency generally classified as AI as of 2017 include successfully understanding man’s speech, challenging at the greatest level in strategic game systems, autonomous cars, intelligent routing in content delivery network and military counterfeiting. The field started on the claim that human intelligence "can be so minutely demonstrated that a machine can be made to counterfeit it". This raises philosophical arguments about the nature of the mind and the ethics of creating artificial beings endowed with human-like intelligence which are issues that have been explored by illusion, fiction and philosophy since the end. Some people also consider AI to be an emergency to humanity if it progresses unabatedly. Others believe that AI, unlike previous technological revolutions, will create a risk of mass unemployment. In the twenty-first century, AI techniques have accomplished a resurgence following circumstantial advances in computer power, large amounts of data, and theoretical understanding; and AI techniques have become an essential part of the technology industry, helping to clarify many confronting problems in computer science.

Phrase History

Artificial intelligence was developed as an academic discipline in 1956, and in the years since has accomplished several waves of anticipation, followed by disappointment and the loss of funding. The study of mechanical or "formal" reasoning began with philosophers and mathematicians in antiquity. The study of mathematical logic led directly to Alan Turing's theory of computation, which advised that a machine, by shuffling symbols as simple as "0" and "1", could simulate any conceivable act of mathematical deduction. This acumen, that digital computers can counterfeit any process of formal reasoning, is known as the Church–Turing thesis. Along with concurrent discoveries in neurobiology, information theory and cybernetics, this led researchers to contemplate the possibility of building an electronic brain. Turing expected that "if a human could not distinguish between replies from a machine and a human, the machine could be considered “intelligent". The first work that is now generally recognized as AI was McCullouch and Pitts' 1943 formal design for Turing-complete "artificial neurons". According to Bloomberg's Jack Clark, in the year 2015 was a landmark year for artificial intelligence, with the number of software projects that use AI within Google increased from a "sporadic usage" in 2012 to more than 2,700 projects. Clark also presents factual data indicating that error rates in image processing tasks have fallen significantly since 2011. He aspect this to an increase in affordable neural networks, due to a rise in cloud computing infrastructure and to an increase in research tools and data sets. Other cited examples include Microsoft's development of a Skype system that can automatically translate from one language to another and Facebook's system that can describe images to blind people.

Continued in the next blog...

Robotics: A Hot Cake of Today's Technology

Image for representative purpose only.

All you need to know about one of the greatest advancement in modern technology: Robotics

Robotics is an integrative  bureau of engineering and science that combines mechanical engineering, electronics engineering, computer science, and others. Robotics comprises of design, construction, operation, and use of robots, as well as computer systems for their control, sensory feedback, and information processing. Computers have transmogrify play, transportation, work, health and medicine, entertainments and sports. Yet for all their power, these machines still cannot achieve a simple tasks that a child can do, such as navigating an unknown room or using a pencil. The idea of creating machines that can serve autonomously dates back to classical times, but research into the functionality and potential uses of robots did not grow essentially until the 20th century. Throughout history, it has been intermittently assumed that robots will one day be able to copycat human behavior and manage tasks in a human-like fashion. Today, robotics is a rapidly growing field, as technological advances continue; researching, designing, and building new robots serve various practical purposes, whether domestically, commercially, or militarily. Many robots are developed to do jobs that are hazardous to people such as defusing bombs, finding survivors in unstable ruins, and exploring mines and shipwrecks. Robotics is also used in STEM (science, technology, engineering, and mathematics) as a teaching aid. The clarification is finally coming within reach. It will emerge from the intersection of two major pursuits: the reverse engineering of the brain and the burgeoning field of artificial intelligence. Over the next 20 years, these two pursuance will combine to usher in a new epoch of intelligent machines.

Phase History

A writer from Czech named Karel Capek in his play Rossum’s Universal Robot has first introduced the word Robot. The word robot comes from the Slavic word rabota, which means labour/work. The play begins in a factory that makes artificial people called robots, creatures who can be mistaken for humans – very similar to the modern ideas of androids. Karel Čapek himself did not coin the word. In the year 1942, a science fiction writer  Isaac Asimov developed his Three Laws of Robotics, later on 1948, Norbert Wiener formulated the principles of cybernetics, the basis of practical robotics. Commercial and industrial robots are outspread today and used to perform jobs more economically, more accurately and more reliably, than humans. They are also employed in some jobs which are too dirty, dangerous, or dull to be suitable for humans. Robots are widely used in developing, manufacturing, assembly, packing and packaging, mining, transport, earth and space exploration, surgery, weaponry, laboratory research, safety, and the mass production of consumer and industrial goods. 

Applications

As more and more robots are designed for specific tasks this method of classification becomes more relevant. For example, many robots are designed for assembly work, which may not be readily adaptable for other applications. They are termed as "assembly robots". In recent years, neuroscientists have learned some remarkable things about the dendrite. One is that each of its branches acts as a set of pattern detectors. Neuroscientists used to believe that learning occurred solely by modifying the effectiveness of existing synapses so that when an input arrived at a synapse it would either be more likely or less likely to make the cell fire. Up to 40 percent of the synapses on a neuron are replaced with new ones every day. New synapses result in new patterns of connections among neurons, and therefore new memories. we can learn new things without interfering with old memories and why we don’t have to retrain the brain every time we learn something new. Some robots are specifically designed for heavy load manipulation, and are labelled as "heavy duty robots". Robots are increasingly used in manufacturing (since the 1960s). In the auto industry, they can amount for more than half of the "labor". There are even "lights off" factories such as an IBM keyboard manufacturing factory in Texas that is 100% automated. Robots can serve as waiters and cooks,  also at home. Boris is a robot that can load a dishwasher. Rotimatic  is a robotics kitchen appliance that cooks flatbreads automatically. Robots all have some kind of mechanical construction, a frame, form or shape designed to achieve a particular task. For example, a robot designed to travel across heavy dirt or mud, might use caterpillar tracks. Intelligent machines don’t have to model all the complexity of biological neurons, but the capabilities enabled by dendrites and learning by rewiring are essential. These capabilities will need to be in future AI systems. All robots contain some level of computer programming code. A program is how a robot decides when or how to do something. In the caterpillar track example, a robot that needs to move across a muddy road may have the correct mechanical construction and receive the correct amount of power from its battery, but would not go anywhere without a program telling it to move. 

Know About Digital Twin

Image for representative purpose only.

A New Benchmark of Digitalization-Digital Twin

Digital twin refers to a digital clone of natural or real equity (physical twin), processes and systems that can be used for wide varieties. The digital depiction gives both the elements and the dynamics of how an Internet of Things device operates and lives all over its life cycle. Digital twins accommodate artificial intelligence, machine learning and software analytics with data to produce living digital simulation models that amend and change as their physical doppelganger change. A digital twin continuously learns and updates itself from multiple sources to represent its near real-time status, working condition or position. This learning applications learns from itself, using sensor data that transfer various aspects of its operating condition; from human experts, such as engineers with expansive and significant industry domain knowledge; from other similar machines; from other similar squadron of machines; and from the larger systems and environment in which it may be a part of. A digital twin also harmonize historical data from past machine control to factor into its digital model. In various financial sectors, twins are being used to forward the action and maintenance of physical assets, systems and manufacturing processes. They are a formative technology for the Industrial Internet of Things, where real objects can animate and interact with other machines and people virtually. With the help of Digital Twin, the world can experience the next big tech leap that is the traversing of real assets with the intelligent pragmatic models. The physical interface has taken its place at the cutting edge of what many are calling the “Fourth Industrial Revolution”. 

There are various video games that we have encountered in our life. These are the part of modern digitalization. The emerging of E-books, then transformation of CD players to DVD and then to MP3 files are the emergence of digital revolution. It solely expresses us and our behavior in a digital environment. Digital twinning involves aligning out a real entity, object, process, or system, and developing an evolving virtual model. By developing smart sensors in the real entity, you can gather real-time data or information. That information then records the state of the digital twin, facilitating it to act as a remotely accessible proxy.

Benefits of Digital Twins

• Smoothing Product Design and Production: Producers discern the cost that goes into scheming and producing a prototype. We could go through various repetitions of prototypes before coming up with a working version. It’s a long, costly process one war widowing on obsolescence. What many companies do now is, before they even start on a physical prototype, they create the digital twin. This virtual mock-up can undergo simulations and analyze to provide feedback that raises the eventual physical prototype. Because the virtual mock-up exists in a digital environment, it’s also easier for geographically versatile teams to collude on the design without meeting in person.
• Predictive Management Analysis:The collected information used to create digital twins can predict when equipment maintenance is needed and even when breakdowns will occur. Timely reporting of these needs to human monitors can save companies both time and money, and deducts downtime for valuable repairs.
• Duplication for Answer Key Questions: Digital twins’ addition extends good way off a product’s design phase and its natural lifecycle. The twin can also reply vital questions about how an entity will respond to stresses and various conditions. Then you could predict the information and manipulate what the impact would be to the real copy. Seeing how this component and other smart entity, might react individually and collectively can influence near future operational and strategic decisions.

One entity may not seem necessary, but given the contingency to produce a digital version of a business' full set of assets and equipment, Digital Twins can provide important, actionable acumen into operations and supply chains. A little bit than tinker with real equipment, for example, data pulled from IoT sensors can be used to produce a virtual replica for the purpose of real-time checking and validation. Digital Twins can improve businesses and discover methods to advance operations, design, and test products -- before fully investing in them -- without causing interruption or failures in the supply chain by tampering with real time entity.48 percent of organization players which are already deploying IoT are using or plan to use Digital Twin technologies by the end of 2018. The deployment of Digital Twin technologies is not a small project and must be examined in addition with the return on likely high levels of investment. However, according to  report issued by Deloitte, the technology is quickly growing across industries including aerospace, retail, healthcare, and so on.

Amazon Alexa Can Cause Major Threat

Image for representative purpose only.

Amazon's Alexa-powered Echo device recorded private conversation and sent it to acquainted person without user consent

Amazon, the biggest online marketing company has a brand of smart speakers called 'Amazon Echo' that connects to an intelligent command-controlled personal assistant service known as 'Alexa'. This is a kind of similar functionality like Google Assistant. The word 'Alexa' is like the activating keyword for the system to enable its functionalities. Different functionalities of this device comprises of interaction and operations using voice commands, setting alarms, playing musics, making to-do list, weather forecast, podcasts streaming, traffic and other real-time information. This device acts as an automation hub for home as it can connect and control several other smart devices. But such amazing smart device can sometimes cause serious problems and threats as their artificial intelligence system cannot completely match with human intelligence.

Recently, a woman in Oregon's largest city Portland reported that she received a phone call from one of her husband's employee claiming that he received a audio file containing personal conversation of their house regarding hardwood floors. This is really shocking and an invasion of privacy for the woman whose entire house is connected with this device. Amazon also confirmed the mishap that the private conversation had been unintentionally recorded by the device and send to random person of the woman's contact list. As explained by the company, the device may have activated itself from word in background conversation which it has interpreted as 'Alexa' and then maybe some another words which it has interpreted as request for recording and sending the message. The 'Alexa' may have responded asking for contact name and it has again interpreted the background conversation with a name similar to someone in the contact list of the women. It may have got confirmation again in same manner after which the recorded audio has been sent. This incident reflects that there is immense need of improvement and training in the artificial intelligence systems of Amazon or these type of incidents may cause serious threat to someone in future.

Google Assistant

Image for representative purpose only.

Google Assistant an unbelievable achievement in the world of Artificial Intelligence

What is Google Assistant?

The Google Assistant is a virtual assistant developed by Google that is primarily available on mobile and smart home devices.Google's voice-controlled smart assistant.It's was originally an extension of Google Now - designed to be personal, while enlarging on Google's existing "OK Google" voice controls. Unlike Google Now, the Google Assistant can engage in two-way conversations.In May 2016, Google Assistant initially released as part of Google's messaging app Allo and its voice-enabled speaker Google Home. Later in the beginning of 2017 in the month of February, it began to be set up on other Android devices including third-party smart phones and Android Wear. In May 2017, it was released as a standalone application in the IOS operating system. The Google Assistant has been further enhanced and enlarged in various devices including cars and smart home appliances.

How does Google Assistant Works?

End-users primarily interact with the Google Assistant through natural voice, though keyboard input is also backed. In the same nature and way as Google Now, the Google Assistant is capable to explore the Internet, schedule events and alarms, reconcile hardware settings on the user's device, and show relevant data from the user's Google account. Google also announced that the Assistant will able to recognize objects and collect visual information through device’s camera. It can also purchase products and send money, as well as identify music. Google Assistant knows us and understands context, it will react in a smart way. Assistant spreads its wings, because it gives voice control a lot more power and moves it on from only reacting to specific commands. Google Assistant gets 6 new voices, including John Legend's. We will now be able to choose among six new voices for Google Assistant for our Android phone or Google Home. Google even say that, in future Assistant will able to call and book appointments for us. Google Assistant is now in everything and can control anything, with full list of Google Assistant partners.

Which devices offer Google Assistant?

Google Assistant launched on the Google Pixel smart phones and Google Home in 2016, before expanding to just about all modern Android devices.  The supported devices are Google Home Max, Android Lollipop and higher (Android 5.0+) phones (with 1.4 GB RAM or higher and 720p (HD) resolution or higher) , Android Lollipop and higher (Android 5.0+) tablets, Android Go phones iOS 10 and higher iPhones and iPads (through a free app),Wear OS 2.0+ smartwatches,  Google Pixelbook, Nokia 8110 4G, Jio Phone, JBL Link 300,Sony LF-S50G, Zolo Mojo (from Anker), Panasonic GA10, Onkyo G3, Mobvoi TicHome Mini, Insignia Smart Voice Speaker, Nvidia Shield set-top box, iHome iGV1 smart clock, LG ThinQ smart speaker, LG Smart TVs (all 2018 models), Nest Cam IQ Indoor, Volvo vehicles (select models with the Sensus infotainment system), LG ThinQ View smart display (coming in summer 2018), Lenovo Smart Display, JBL View smart display ,Sony smart display ,Archos Hello smart displays , Sonos One. 

How do we turn on our Google Assistant?

The five easy steps to turn on Google Assistant are:

• Change "Ok Google" settings
• Turn on our phone or tablet, touch and hold the Home button or say "Ok Google."
• At the top right, tap More Settings.
• Under "Devices," select your phone or tablet.
• Turn on Google Assistant turn “Ok Google” detection on or off. If you don't see these steps, you're using an earlier version of the Google app.

How do I know if my Phone has Google Assistant?

To check if our phone has Google Assistant, say "Ok Google" or press-and-hold the home button. On Marshmallow and Nougat devices that have the updated version, that long press will launch Assistant, popping up with a page asking the way it can help. That is the initial point for Assistant, after which we can type or give our voice and have Assistant respond.

Criticism of Google Assistant

After the announcement, concerns were made over the moralistic and community questions that artificial intelligence technology such as Duplex raises. For example, manmade operators may not notice that they are speaking with a digital robot when conversing with Duplex, which some expert opinion as unethical. Concerns over confidentiality were also spotted, as conversations with Duplex are recorded in order for the virtual assistant to analyze and respond. Google mentioned that, in certain laws, the assistant would inform those on the other end of the phone that the call is being recorded.

Facebook Checking on Fake Accounts

Image for representative purpose only.

Facebook is checking its billion user accounts and removing fake accounts

After the Cambridge Ananlytica scandal, the social giant Facebook is continuously updating its security and privacy policies to restore their faith on their billion user base. After the data leak, Facebook made public its secret community standard policies through which they provided user with the flexibility to control data sharing and the types of content sharing which do not violate Facebook's policy. The internal team and the artificial intelligence systems are working hard to check the contents posted on Facebook to prevent the spread of hatred, discrimination, violence and terrorism. Now again in another new move Facebook is checking and removing huge number of fake accounts from its social platform. It has already removed 583 million fake accounts out of its billion user base within the first quarter of 2018 to prevent spamming, privacy scandals and misuse of the social platform for violence, hatred and discrimination. The Facebook's artificial intelligence systems are really working effectively to prevent spams, bots and fake accounts and all the fake accounts are deleted within few minutes of registration. But Facebook thinks that their artificial intelligence system is not full proof and still they may have very less percent of fake accounts existing. The artificial intelligence systems have been deleting millions of spam post, fake accounts, violent and nudity content even before giving chance to user to report them. Facebook is continuously upgrading its artificial intelligence systems for more better detection of spam posts, fake accounts and hate speeches in order to restore faith of common people over social media. These active steps will prevent Facebook to be used for platform of hatred, discrimination, violence and terrorism and thus can again restore the faith of millions of users over social media.