John Launchbury of DARPA has an excellent video that I recommend everyone watch ( viewing just the slides will give one a wrong impression of the content). Statistical Learning -- Where programmers create statistical models for specific problem domains and train them on big data. Contextual Adaptation -- Where systems construct contextual explanatory models for classes of real world phenomena. It's a bit of a simplified presentation because it lumps all of machine learning, Bayesian methods and Deep Learning into a single category. There are many more approaches to AI that don't fit within DARPA's 3 waves.

Artificial intelligence (AI) promises to revolutionise our lives, drive our cars, diagnose our health problems, and lead us into a new future where thinking machines do things that we're yet to imagine. Even billionaire entrepreneur Elon Musk, who admits he has access to some of the most cutting-edge AI, said recently that without some regulation "AI is a fundamental risk to the existence of human civilization". So what is the future of AI? Michael Milford and Peter Stratton are both heavily involved in AI research and they have different views on how it will impact on our lives in the future. Answering this question depends on what you consider to be "artificial intelligence". Basic machine learning algorithms underpin many technologies that we interact with in our everyday lives - voice recognition, face recognition - but are application-specific and can only do one very specific defined task (and not always well).

Sequence prediction is a problem that involves using historical sequence information to predict the next value or values in the sequence. The sequence may be symbols like letters in a sentence or real values like those in a time series of prices. Sequence prediction may be easiest to understand in the context of time series forecasting as the problem is already generally understood. In this post, you will discover the standard sequence prediction models that you can use to frame your own sequence prediction problems. Recurrent Neural Networks, like Long Short-Term Memory (LSTM) networks, are designed for sequence prediction problems.

Generating adversarial examples is a critical step for evaluating and improving the robustness of learning machines. So far, most existing methods only work for classification and are not designed to alter the true performance measure of the problem at hand. We introduce a novel flexible approach named Houdini for generating adversarial examples specifically tailored for the final performance measure of the task considered, be it combinatorial and non-decomposable. We successfully apply Houdini to a range of applications such as speech recognition, pose estimation and semantic segmentation. In all cases, the attacks based on Houdini achieve higher success rate than those based on the traditional surrogates used to train the models while using a less perceptible adversarial perturbation.

Modeling physiological time-series in ICU is of high clinical importance. However, data collected within ICU are irregular in time and often contain missing measurements. Since absence of a measure would signify its lack of importance, the missingness is indeed informative and might reflect the decision making by the clinician. Here we propose a deep learning architecture that can effectively handle these challenges for predicting ICU mortality outcomes. The model is based on Long Short-Term Memory, and has layered attention mechanisms. At the sensing layer, the model decides whether to observe and incorporate parts of the current measurements. At the reasoning layer, evidences across time steps are weighted and combined. The model is evaluated on the PhysioNet 2012 dataset showing competitive and interpretable results.

Yann LeCun, arguably the father of modern machine learning, has described Generative Adversarial Networks (GANs) as the most interesting idea in deep learning in the last 10 years (and there have been a lot of interesting ideas in Machine Learning over the past 10 years). From a trading perspective, or any game theoretic activity in which the game itself constantly evolves, GANs are big news. This is wrong, because it doesn't address the evolution of the game and the other players. It also comes up with all kinds of nonsense based on spurious correlations. Bigger data set more false positives.

You might not know it, but machine learning already plays a part in your everyday life. When you speak to your phone (via Cortana, Siri or Google Now) and it fetches information, or you type in the Google search box and it predicts what you are looking for before you finish, you are doing something that has only been made possible by machine learning. However, this is just the beginning: with companies such as Google, Microsoft and Facebook spending millions on research into advanced neural networks and deep machine learning, computers are set to get smarter still. This is a story about how ingenious algorithms and code are giving computers the ability to do things we never previously thought possible. Machine learning and deep learning have grown from the same roots within computer science, using many of the same concepts and techniques.

The team at Alphabet have used a reinforced learning program to teach the DeepMind AI how to do parkour. Reinforced learning (RL) is a common tool for teaching and guiding behavior by using a reward system. Basically good or desirable behavior gets rewards and undesirable behavior gets nothing. The aim of the project was to investigate if simple rewards systems would also work in complex environments. A virtual parkour course was designed with steps, ledges, hurdles, and drops.

For some AI problems, such as playing Atari or Go, the goal is easy to define - it's winning. But how do you describe the process for performing a backflip? The difficulty of accurately describing a complex behaviour is a common problem when teaching motor skills to an artificial system. In this work we explore how sophisticated behaviors can emerge from scratch from the body interacting with the environment using only simple high-level objectives, such as moving forward without falling. Specifically, we trained agents with a variety of simulated bodies to make progress across diverse terrains, which require jumping, turning and crouching.

Sebastiaan Vaessen, head of strategy at Naspers warns those business leaders who have only just started thinking of being digital or being mobile-first, are probably already behind the curve – they will need to start thinking about how to become'AI-first'. Vaessen said that artificial intelligence had a major breakthrough in 2016 when Google DeepMind's AlphaGo computer programme defeated the world GO champion. This achievement, he said, proved that AI is now able to complete complex tasks that would take people years to master. He said that AI has already mastered a remarkable list of human tasks, ranging from translations to more creative undertakings like writing music and film scripts. Vaessen outlined the potential and relevance of AI for the world of business.