"Many researchers … speculate that the information-processing abilities of biological neural systems must follow from highly parallel processes operating on representations that are distributed over many neurons. [Artificial neural networks] capture this kind of highly parallel computation based on distributed representations"
– from Machine Learning (Section 4.1.1; page 82) by Tom M. Mitchell, McGraw Hill Companies, Inc. (1997).
Deep reinforcement learning (DRL) is transitioning from a research field focused on game playing to a technology with real-world applications. Notable examples include DeepMind's work on controlling a nuclear reactor or on improving Youtube video compression, or Tesla attempting to use a method inspired by MuZero for autonomous vehicle behavior planning. But the exciting potential for real world applications of RL should also come with a healthy dose of caution – for example RL policies are well known to be vulnerable to exploitation, and methods for safe and robust policy development are an active area of research. At the same time as the emergence of powerful RL systems in the real world, the public and researchers are expressing an increased appetite for fair, aligned, and safe machine learning systems. The focus of these research efforts to date has been to account for shortcomings of datasets or supervised learning practices that can harm individuals.
Tiernan Ray has been covering technology and business for 27 years. He was most recently technology editor for Barron's where he wrote daily market coverage for the Tech Trader blog and wrote the weekly print column of that name. DeepMind's "Gato" neural network excels at numerous tasks including controlling robotic arms that stack blocks, playing Atari 2600 games, and captioning images. The world is used to seeing headlines about the latest breakthrough by deep learning forms of artificial intelligence. The latest achievement of the DeepMind division of Google, however, might be summarized as, "One AI program that does a so-so job at a lot of things."
According to Gartner, AI applies advanced analysis and logic-based techniques, including machine learning, to interpret events, support and automate decision-making, and take action. In essence, the concept of AI centres on enabling computer systems to think and act in a more'human' way, by learning from and responding to the vast amounts of information they're able to use. AI is already transforming our everyday lives. From the AI features on our smartphones such as built-in smart assistants, to the AI-curated content and recommendations on our social media feeds and streaming services. As the name suggests, machine learning is based on the idea that systems can learn from data to automate and improve how things are done – by using advanced algorithms (a set of rules or instructions) to analyse data, identify patterns and make decisions and recommendations based on what they find.
Machine learning and other artificial intelligence (AI) methods have had immense success with scientific and technical tasks such as predicting how protein molecules fold and recognising faces in a crowd. However, the application of these methods to the humanities is yet to be fully explored. What can AI tell us about philosophy and religion, for example? As a starting point for such an exploration, we used deep learning AI methods to analyse English translations of the Bhagavad Gita, an ancient Hindu text written originally in Sanskrit. Using a deep learning-based language model called BERT, we studied sentiment (emotions) and semantics (meanings) in the translations.
Tensorflow is an open-source end-to-end machine learning framework that makes it easy to train and deploy the model. It consists of two words - tensor and flow. A tensor is a vector or a multidimensional array that is a standard way of representing the data in deep learning models. Flow implies how the data moves through a graph by undergoing the operations called nodes. It is used for numerical computation and large-scale machine learning by bundling various algorithms together.
I first came across the legend of Hinton in a fabulous book by Cade Metz called Genius Makers, where he detailed the lives of those who shaped AI, foremost among them being Hinton. After studying psychology at Cambridge and AI at the University of Edinburgh, Hinton went back to something which had fascinated him even as a child: How the human brain stored memories, and how it worked. He was one of the first researchers who started working on'mimicking' the human brain with computer hardware and software, thus constructing a newer and purer form of AI, which we now call'deep learning'. He started doing this in the 1980s, along with an intrepid bunch of students. His PhD thesis, titled Deep Neural Networks for Acoustic Modelling in Speech Recognition, demonstrated how deep neural networks outclassed older machine learning models like Hidden Markovs and Gaussian Mixtures at identifying speech patterns.
Image segmentation is an aspect of computer vision that deals with segmenting the contents of objects visualized by a computer into different categories for better analysis. The contributions of image segmentation in solving a lot of computer vision problems such as analysis of medical images, background editing, vision in self driving cars and analysis of satellite images make it an invaluable field in computer vision. One of the greatest challenges in computer vision is keeping the space between accuracy and speed performance for real time applications. In the field of computer vision there is this dilemma of a computer vision solution either being more accurate and slow or less accurate and faster. PixelLib Library is a library created to allow easy integration of object segmentation in images and videos using few lines of python code.
Inspired by progress in large-scale language modelling, we apply a similar approach towards building a single generalist agent beyond the realm of text outputs. The agent, which we refer to as Gato, works as a multi-modal, multi-task, multi-embodiment generalist policy. The same network with the same weights can play Atari, caption images, chat, stack blocks with a real robot arm and much more, deciding based on its context whether to output text, joint torques, button presses, or other tokens. During the training phase of Gato, data from different tasks and modalities are serialised into a flat sequence of tokens, batched, and processed by a transformer neural network similar to a large language model. The loss is masked so that Gato only predicts action and text targets.