San Diego, Calif., Oct. 24, 2018 -- A team of U.S. computer scientists is receiving a $10 million grant from the National Science Foundation to make machine learning more secure. The grant establishes the Center for Trustworthy Machine Learning at a consortium of seven universities, including the University of California San Diego. Researchers will work together toward two goals: understanding the risks inherent to machine learning; and developing the tools, metrics and methods to manage and mitigate these risks. The science and arsenal of defensive techniques emerging within the center will provide the basis for building more trustworthy and secure systems in the future, as well as fostering a long-term research community within this essential domain of technology, researchers said. "This research is important because machine learning is becoming more pervasive in our daily lives, powering technologies we interact with, including services like e-commerce and Internet searches, as well as devices such as Internet-connected smart speakers," said Kamalika Chaudhuri, a computer science professor at the Jacobs School of Engineering, who will be leading the UC San Diego portion of the research.

The fundamentals issues in Deep Learning are access to data, processing power and data scientists (i.e. But there are four more fundamental things that set companies apart. In this 10min presentation you will learn what the main challenges in Deep Learning are. For the next part of the tutorial, see a live example of running a TensorFlow MNIST example on Valohai: https://www.youtube.com/watch?v L5CcJ...

According to Deloitte's second State of the AI in the Enterprise report, one of the greatest concerns about artificial intelligence, a technology that is dominating hiring processes for 32 percent of responding firms, is the difficulty in hiring qualified staff to manage it. Of the 1,100 AI-early-adopter executives who participated in the just-released survey, 69 percent were facing an AI skills gap either moderate, major or extreme. Where staff was most lacking were for the roles of AI researchers to create algorithms, software developers to creates systems, and data scientists to manage and analyze the information that fuels the entire AI process. Most agreed that job descriptions will be significantly altered by the growth of AI. In fact, 72 percent had already seen substantial changes in skills and job roles.

Dr Faisal added: "The explosion in Artificial Intelligence applications in healthcare is currently focused on mimicking the perceptual ability of human doctors, e.g. However, doctors do more than just diagnose, they treat people. Our AI Clinician system focuses on capturing this cognitive capacity of doctors: Imagine having a doctor watching over you every second of every day, administering a course of treatment, observing how you respond to the treatment, and then adjusting the treatment as your condition evolves.

Algorithmic bias, or the notion that human biases can be magnified when consciously or subconsciously programmed into algorithms, has been a hot topic in machine learning. How do we create "fair" algorithms that behave in as unbiased a manner as possible? Google has released a free 60-minute online course on fairness as part of its popular Machine Learning Crash Course. This includes a short video lecture, materials on types of bias and how to identify and evaluate for bias, and a programming exercise to put your learning into action. Even if you are not a software engineer, you are a consumer of the fruits of their work and it behoves all of us to educate ourselves on how they function (and malfunction).

University of Cambridge researchers used machine learning to attempt to define why children struggle in school. Scientists at the University of Cambridge in the U.K. have used machine learning to identify why children struggle at school, through an analysis of 550 students having difficulty. The team fed their algorithm cognitive testing information from each child, which demonstrated how the children best fit within four clusters of problems. Two clusters identified working memory skills and processing sounds in words as problem areas, while the other two defined broad cognitive difficulties in many areas and typical cognitive tests for the subjects' age. Cambridge's Joni Holmes says the results suggest "children who are finding the same subjects difficult could be struggling for very different reasons, which has important implications for selecting appropriate interventions."

Recent developments such as Virtual and Augmented Reality as well as the introduction of gamification in corporate learning are changing the face of training. The challenge still remains to engage and entertain as well as teach in an environment that is harder to control by L&D professionals. With corporate education becoming almost entirely learner-centric, the solution I was advocating in one of my previous articles is personalized adaptive learning. It makes sense in the context of things and studies already support its benefits. However, there is some effort to be put in development and implementation so companies may not jump at the idea.

Noname manuscript No. (will be inserted by the editor) Abstract Ensemble techniques are powerful approaches that combine several weak learners to build a stronger one. As a meta-learning framework, ensemble techniques can easily be applied to many machine learning methods. Inspired by ensemble techniques, in this paper we propose an ensemble loss functions applied to a simple regressor. We then propose a half-quadratic learning algorithm in order to find the parameter of the regressor and the optimal weights associated with each loss function. Moreover, we show that our proposed loss function is robust in noisy environments. For a particular class of loss functions, we show that our proposed ensemble loss function is Bayes consistent and robust. Experimental evaluations on several data sets demonstrate that the our proposed ensemble loss function significantly improves the performance of a simple regressor in comparison with state-of-the-art methods. Keywords Loss function · Ensemble methods · Bayes Consistent Loss function · Robustness 1 Introduction Loss functions are fundamental components of machine learning systems and are used to train the parameters of the learner model.

In this paper, we study online convex optimization in dynamic environments, and aim to bound the dynamic regret with respect to any sequence of comparators. Existing work have shown that online gradient descent enjoys an $O(\sqrt{T}(1+P_T))$ dynamic regret, where $T$ is the number of iterations and $P_T$ is the path-length of the comparator sequence. However, this result is unsatisfactory, as there exists a large gap from the $\Omega(\sqrt{T(1+P_T)})$ lower bound established in our paper. To address this limitation, we develop a novel online method, namely adaptive learning for dynamic environment (Ader), which achieves an optimal $O(\sqrt{T(1+P_T)})$ dynamic regret. The basic idea is to maintain a set of experts, each attaining an optimal dynamic regret for a specific path-length, and combines them with an expert-tracking algorithm. Furthermore, we propose an improved Ader based on the surrogate loss, and in this way the number of gradient evaluations per round is reduced from $O(\log T)$ to $1$. Finally, we extend Ader to the setting that a sequence of dynamical models is available to characterize the comparators.

We present a modular approach for learning policies for navigation over long planning horizons from language input. Our hierarchical policy operates at multiple timescales, where the higher-level master policy proposes subgoals to be executed by specialized sub-policies. Our choice of subgoals is compositional and semantic, i.e. they can be sequentially combined in arbitrary orderings, and assume human-interpretable descriptions (e.g. 'exit room', 'find kitchen', 'find refrigerator', etc.). We use imitation learning to warm-start policies at each level of the hierarchy, dramatically increasing sample efficiency, followed by reinforcement learning. Independent reinforcement learning at each level of hierarchy enables sub-policies to adapt to consequences of their actions and recover from errors. Subsequent joint hierarchical training enables the master policy to adapt to the sub-policies.