Avinash C. Pillai

                                                                                       Neural Networks An artificial neural network is a collection of smaller units called neurons, which are computing units modeled on the way the human brain processes information.  Artificial neural networks borrow some ideas from the biological neural network of the brain, in order to approximate some of its processing results. These units or neurons take incoming data like the biological neural networks and learn to make decisions over time. Neural networks learn through a process called backpropagation. Backpropagation uses a set of training data that match known inputs to desired outputs. First, the inputs are plugged into the network and outputs are determined. Then, an error function determines how far the given output is from the desired output. Finally, adjustments are made in order to reduce errors. A collection of neurons is called a layer, and a layer takes in an input and provides an output

                                     Deep Learning While Machine Learning is a subset of Artificial Intelligence ,  Deep Learning is a specialized subset of Machine Learning .  Deep Learning layers algorithms to create a Neural Network, an artificial replication of the structure and functionality of the brain, enabling AI systems to continuously learn on the job and improve the quality and accuracy of results. This is what enables these systems to learn from unstructured data such as photos, videos, and audio files. Deep Learning, for example, enables natural language understanding capabilities of AI systems, and allows them to work out the context and intent of what is being conveyed. Deep learning algorithms do not directly map input to output. Instead, they rely on several layers of processing units. Each layer passes its output to the next layer, which processes it and passes it to the next. The many layers is why it’s called deep learning.  When creating deep learning algorithms,

                      Machine Learning Techniques and Training Machine Learning is a broad field and we can split it up into three different categories, Supervised Learning, Unsupervised Learning, and Reinforcement Learning. There are many different tasks we can solve with these.  Supervised Learning refers to when we have class labels in the dataset and we use these to build the classification model. What this means is when we receive data, it has labels that say what the data represents. In a previous example, we had a table with labels such as age or sex.  With Unsupervised Learning, we don't have class labels and we must discover class labels from unstructured data. This could involve things such as deep learning looking at pictures to train models. Things like this are typically done with something called clustering. Reinforcement Learning is a different subset, and what this does is it uses a reward function to penalize bad actions or reward good actions. Breaking down Supe

                                                    Machine Learning Machine Learning, a subset of AI, uses computer algorithms to analyze data and make intelligent decisions based on what it has learned. Instead of following rules-based algorithms, machine learning builds models to classify and make predictions from data. Let's understand this by exploring a problem we may be able to tackle with Machine Learning.  What if we want to determine whether a heart can fail, is this something we can solve with Machine Learning? The answer is, Yes. Let's say we are given data such as beats per minute, body mass index, age, sex, and the result whether the heart has failed or not. With Machine Learning given this dataset, we are able to learn and create a model that given inputs, will predict results. So what is the difference between this and using statistical analysis to create an algorithm? An algorithm is a mathematical technique. With traditional programming, we take data and rules

                      Terminology and Related Concepts Let's differentiate some of the closely related terms and concepts of AI:  Artificial intelligence. Machine learning. Deep learning. Neural networks.  These terms are sometimes used interchangeably, but they do not refer to the same thing.  Artificial intelligence is a branch of computer science dealing with a simulation of intelligent behavior. AI systems will typically demonstrate behaviors associated with human intelligence such as planning, learning, reasoning, problem-solving, knowledge representation, perception, motion, and manipulation, and to a lesser extent social intelligence and creativity.  Machine learning is a subset of AI that uses computer algorithms to analyze data and make intelligent decisions based on what it has learned, without being explicitly programmed. Machine learning algorithms are trained with large sets of data, and they learn from examples. They do not follow rules-based algorithms.  Machine lea

                           Cognitive Computing                 (Perception, Learning, Reasoning) AI is at the forefront of a new era of computing, Cognitive Computing. It's a radically new kind of computing, very different from the programmable systems that preceded it, as different as those systems were from the tabulating machines of a century ago.  Conventional computing solutions, based on the mathematical principles that emanate from the 1940's, are programmed based on rules and logic intended to derive mathematically precise answers, often following a rigid decision tree approach. But with today's wealth of big data and the need for more complex evidence-based decisions, such a rigid approach often breaks or fails to keep up with available information. Cognitive Computing enables people to create a profoundly new kind of value, finding answers and insights locked away in volumes of data. Whether we consider a doctor diagnosing a patient, a wealth manager advising a cl

               Generative AI Applications Generative AI has emerged as a powerful technology that enables software applications to create, generate, and simulate new content, enhancing their capabilities and providing unique experiences. Unlike traditional software that follows predefined rules and algorithms, generative AI leverages machine learning and deep learning techniques to learn patterns and generate original content based on the knowledge it has acquired during training.  Due to its potential to create new, personalized content that would have been impossible to create otherwise, Generative AI has been used in various fields, leading to the development of numerous engaging and well-liked applications. Some popular applications of Generative AI in action include:  1: Generative Pre-trained Transformers or GPT , is a family of large language models developed by OpenAI that are capable of producing human-like text. GPT-3.5 and GPT-4 are iterations in this family of models with m