Recent advances in deep reinforcement learning have made significant strides in performance on applications such as Go and Atari games. However, developing practical methods to balance exploration and exploitation in complex domains remains largely unsolved. Thompson Sampling and its extension to reinforcement learning provide an elegant approach to exploration that only requires access to posterior samples of the model. At the same time, advances in approximate Bayesian methods have made posterior approximation for flexible neural network models practical. Thus, it is attractive to consider approximate Bayesian neural networks in a Thompson Sampling framework. To understand the impact of using an approximate posterior on Thompson Sampling, we benchmark well-established and recently developed methods for approximate posterior sampling combined with Thompson Sampling over a series of contextual bandit problems. We found that many approaches that have been successful in the supervised learning setting underperformed in the sequential decision-making scenario. In particular, we highlight the challenge of adapting slowly converging uncertainty estimates to the online setting.

Residual Networks (ResNets) have become state-of-the-art models in deep learning and several theoretical studies have been devoted to understanding why ResNet works so well. One attractive viewpoint on ResNet is that it is optimizing the risk in a functional space by combining an ensemble of effective features. In this paper, we adopt this viewpoint to construct a new gradient boosting method, which is known to be very powerful in data analysis. To do so, we formalize the gradient boosting perspective of ResNet mathematically using the notion of functional gradients and propose a new method called ResFGB for classification tasks by leveraging ResNet perception. Two types of generalization guarantees are provided from the optimization perspective: one is the margin bound and the other is the expected risk bound by the sample-splitting technique. Experimental results show superior performance of the proposed method over state-of-the-art methods such as LightGBM.

In this paper, we propose a novel sufficient decrease technique for stochastic variance reduced gradient descent methods such as SVRG and SAGA. In order to make sufficient decrease for stochastic optimization, we design a new sufficient decrease criterion, which yields sufficient decrease versions of stochastic variance reduction algorithms such as SVRG-SD and SAGA-SD as a byproduct. We introduce a coefficient to scale current iterate and to satisfy the sufficient decrease property, which takes the decisions to shrink, expand or even move in the opposite direction, and then give two specific update rules of the coefficient for Lasso and ridge regression. Moreover, we analyze the convergence properties of our algorithms for strongly convex problems, which show that our algorithms attain linear convergence rates. We also provide the convergence guarantees of our algorithms for non-strongly convex problems. Our experimental results further verify that our algorithms achieve significantly better performance than their counterparts.

NEW DELHI: Stressing that the advent of artificial intelligence (AI) will not make jobs disappear, Stephen Wozniak, co-founder of Apple Inc said that on the contrary it will create more jobs.

NEW DELHI: Stressing that the advent of artificial intelligence (AI) will not make jobs disappear, Steve Wozniak, co-founder of Apple Inc said that on the contrary it will create more jobs. Wozniak was speaking at the fourth edition of the Economic Times Global Business Summit at New Delhi on Saturday.

AI is some thing that has been creating buzz all around the world. As a computer engineering graduate, I had have some theoretical knowledge about this subject, but never knew the implementation details about it. Currently, enrolled to Masters in computer engineering and as Junior Lecturer for an IT college, I was having a very busy schedule. Then I heard about this AI workshop being conducted by AI Developers Nepal, I was very excited to implement the thing that I have theoritically known during my college days. So, despite of having a very busy schedule, I decided to get enrolled to the workshop.

The dawn of the artificial intelligence age is upon us and the speed of technological development threatens to leave regulatory control in its wake. Those concerns were one of the key themes to emerge from the World Government Summit staged in Dubai this week. From agriculture and transport to healthcare and education, technology that was once considered science fiction is edging closer to reality. How that is managed in the decades to come is providing an imminent conundrum for governments and policy makers, and proved a common topic of discussion in forums during the three-day summit. "I believe we are right at the start of this revolution happening right now and I imagine more natural ways of communication so it will become seamless," said Carol Riley, president of Drive – AI. "This means non-verbal communication and machines start to understand what we are thinking."

When you think about Artificial Intelligence, it might feel like some far-flung sci-fi fantasy. As usual, reality is stranger than fiction: AI is already practically everywhere. Whether you're ready or not, AI is here and it's here to stay. Governments and companies around the world are already clamouring to implement AI technology. Tractica predicts that AI software revenue will increase from $3.2 billion in 2016 to $89.8 billion by 2025.

Humanity is turning a corner. The signpost is marked "AI." Everyone knows what it stands for. Who knows what it means? We know what it's beginning to look like.

Earlier this week Google and Verily Life Sciences shared the latest advance in computer vision to identify signs of heart disease. With an accuracy of 70 percent, early results from the AI trained on retinal scan images from more than 200,000 patients is as precise as methods that require blood tests for cholesterol, said Google Brain product manager Lily Peng. It's the latest example of AI being used to tackle the biggest killer in the world: heart disease. It takes more lives than any other cause of death -- 800,000 in the United States alone, according to the American Heart Association. To save lives, an AI army is joining the fight.