Welcome to the Cool things this week series. Every week, I'll share a few of papers or articles I've come across. The hope is that others will share some links in the comments section. This category will include breakthroughs that have the potential of enhancing humans or the world we live in. The authors present BrainNet, the first "multi-person non-invasive direct brain-to-brain interface for collaborative problem solving…the interface allows three human subjects to collaborate and solve a task using direct brain-to-brain communication."

Mumbai: Sebastian Thrun is a man of many parts. The president and co-founder of e-learning company Udacity, is not only an innovator and computer scientist but also CEO of Kitty Hawk Corporation that makes flying cars and chairman of Cresta.ai--a Germany-born Thrun was earlier a Google VP and Fellow. At Google, he founded Google X and Google's self-driving car team. He is currently also an Adjunct Professor at Stanford University and at Georgia Tech.

Meta-learning, or learning to learn, is the science of systematically observing how different machine learning approaches perform on a wide range of learning tasks, and then learning from this experience, or meta-data, to learn new tasks much faster than otherwise possible. Not only does this dramatically speed up and improve the design of machine learning pipelines or neural architectures, it also allows us to replace hand-engineered algorithms with novel approaches learned in a data-driven way. In this chapter, we provide an overview of the state of the art in this fascinating and continuously evolving field.

The recent practical successes [26] of Artificial Intelligence (AI) programs of the Reinforcement Learning and Deep Learning varieties in game playing, natural language processing and image classification, are now calling attention to the envisioned pitfalls of their hypothetical extension to wider domains of human behavior. Several voices from the industry and academia are now routinely raising concerns over the advances [49] of often heavily media-covered representatives of this new generation of programs such as Deep Blue, Watson, Google Translate, AlphaGo and AlphaZero. Most of these cutting-edge algorithms generally fall under the class of supervised learning, a branch of the still evolving taxonomy of Machine Learning techniques in AI research. In most cases the implementation choice is artificial neural networks software, the workhorse of the Connectionism school of thought in both AI and Cognitive Psychology. Confronting the current wave of connectionist architectures, critics usually raise issues of interpretability (Can the remarkable predictive capabilities be 1 trusted in real-life tasks? Are these capabilities transferable to unfamiliar situations or to different tasks altogether? How informative are the results about the real world; about human cognition?

We formulate probabilistic numerical approximations to solutions of ordinary differential equations (ODEs) as problems in Gaussian process (GP) regression with non-linear measurement functions. This is achieved by defining the measurement sequence to consists of the observations of the difference between the derivative of the GP and the vector field evaluated at the GP---which are all identically zero at the solution of the ODE. When the GP has a state-space representation, the problem can be reduced to a Bayesian state estimation problem and all widely-used approximations to the Bayesian filtering and smoothing problems become applicable. Furthermore, all previous GP-based ODE solvers, which were formulated in terms of generating synthetic measurements of the vector field, come out as specific approximations. We derive novel solvers, both Gaussian and non-Gaussian, from the Bayesian state estimation problem posed in this paper and compare them with other probabilistic solvers in illustrative experiments.

Motion analysis is used in computer vision to understand the behaviour of moving objects in sequences of images. Optimising the interpretation of dynamic biological systems requires accurate and precise motion tracking as well as efficient representations of high-dimensional motion trajectories so that these can be used for prediction tasks. Here we use image sequences of the heart, acquired using cardiac magnetic resonance imaging, to create time-resolved three-dimensional segmentations using a fully convolutional network trained on anatomical shape priors. This dense motion model formed the input to a supervised denoising autoencoder (4Dsurvival), which is a hybrid network consisting of an autoencoder that learns a task-specific latent code representation trained on observed outcome data, yielding a latent representation optimised for survival prediction. To handle right-censored survival outcomes, our network used a Cox partial likelihood loss function. In a study of 302 patients the predictive accuracy (quantified by Harrell's C-index) was significantly higher (p < .0001) for our model C=0.73 (95$\%$ CI: 0.68 - 0.78) than the human benchmark of C=0.59 (95$\%$ CI: 0.53 - 0.65). This work demonstrates how a complex computer vision task using high-dimensional medical image data can efficiently predict human survival.

Classical signal recovery based on $\ell_1$ minimization solves the least squares problem with all available measurements via sparsity-promoting regularization. In practice, it is often the case that not all measurements are available or required for recovery. Measurements might be corrupted/missing or they arrive sequentially in streaming fashion. In this paper, we propose a global sparse recovery strategy based on subsets of measurements, named JOBS, in which multiple measurements vectors are generated from the original pool of measurements via bootstrapping, and then a joint-sparse constraint is enforced to ensure support consistency among multiple predictors. The final estimate is obtained by averaging over the $K$ predictors. The performance limits associated with different choices of number of bootstrap samples $L$ and number of estimates $K$ is analyzed theoretically. Simulation results validate some of the theoretical analysis, and show that the proposed method yields state-of-the-art recovery performance, outperforming $\ell_1$ minimization and a few other existing bootstrap-based techniques in the challenging case of low levels of measurements and is preferable over other bagging-based methods in the streaming setting since it performs better with small $K$ and $L$ for data-sets with large sizes.

Advances in renewable energy generation and introduction of the government targets to improve energy efficiency gave rise to a concept of a Zero Energy Building (ZEB). A ZEB is a building whose net energy usage over a year is zero, i.e., its energy use is not larger than its overall renewables generation. A collection of ZEBs forms a Zero Energy Community (ZEC). This paper addresses the problem of energy sharing in such a community. This is different from previously addressed energy sharing between buildings as our focus is on the improvement of community energy status, while traditionally research focused on reducing losses due to transmission and storage, or achieving economic gains. We model this problem in a multi-agent environment and propose a Deep Reinforcement Learning (DRL) based solution. Each building is represented by an intelligent agent that learns over time the appropriate behaviour to share energy. We have evaluated the proposed solution in a multi-agent simulation built using osBrain. Results indicate that with time agents learn to collaborate and learn a policy comparable to the optimal policy, which in turn improves the ZEC's energy status. Buildings with no renewables preferred to request energy from their neighbours rather than from the supply grid.

The power of machine learning systems not only promises great technical progress, but risks societal harm. As a recent example, researchers have shown that popular word embedding algorithms exhibit stereotypical biases, such as gender bias. The widespread use of these algorithms in machine learning systems, from automated translation services to curriculum vitae scanners, can amplify stereotypes in important contexts. Although methods have been developed to measure these biases and alter word embeddings to mitigate their biased representations, there is a lack of understanding in how word embedding bias depends on the training data. In this work, we develop a technique for understanding the origins of bias in word embeddings. Given a word embedding trained on a corpus, our method identifies how perturbing the corpus will affect the bias of the resulting embedding. This can be used to trace the origins of word embedding bias back to the original training documents. Using our method, one can investigate trends in the bias of the underlying corpus and identify subsets of documents whose removal would most reduce bias. We demonstrate our techniques on both a New York Times and Wikipedia corpus and find that our influence function-based approximations are extremely accurate.

Explaining the output of a complicated machine learning model like a deep neural network (DNN) is a central challenge in machine learning. Several proposed local explanation methods address this issue by identifying what dimensions of a single input are most responsible for a DNN's output. The goal of this work is to assess the sensitivity of local explanations to DNN parameter values. Somewhat surprisingly, we find that DNNs with randomly-initialized weights produce explanations that are both visually and quantitatively similar to those produced by DNNs with learned weights. Our conjecture is that this phenomenon occurs because these explanations are dominated by the lower level features of a DNN, and that a DNN's architecture provides a strong prior which significantly affects the representations learned at these lower layers. NOTE: This work is now subsumed by our recent manuscript, Sanity Checks for Saliency Maps (to appear NIPS 2018), where we expand on findings and address concerns raised in Sundararajan et. al. (2018).