Kevin Gray: AI has become part of our daily lives, hasn't it! Dr. Anna Farzindar: I was working on my laptop when my college daughter said "Mom please don't do anything wrong with AI!" Then two days later during our family dinner, my younger freshman high school daughter told a story about a video on social media showing a small home care robot that tricked the owner and lied. She asked me "Mom, aren't you afraid of robots?" These short conversations made me think about how the new generation is a big consumer of technology but, at the same time, they are concerned and worried about the future AI. KG: Getting back to basics, what is AI? AF: From talking to your virtual assistance on smartphone (like SIRI), watching a recommended movie on Netflix, searching on Google, following the suggested Instagram posts, using the sophisticated methods of an auto trading stock market, applying the decision making systems for your loan approval, or (soon) sitting in a self-driving car, AI algorithms are so embedded in our daily life that is hard to imagine living a single day without them!

Andrew Bird is the data and analytics manager for Vesparum Capital. He leads the software and data science teams at Vesparum, overseeing full stack web development in Django / React. He is an Australian actuary (FIAA, CERA), who has previously worked with Deloitte Consulting in financial services. Andrew also currently works as a full-stack developer for Draftable Pvt. Ltd. He manages ongoing development of the donation portal for Effective Altruism Australia website, on a voluntary basis.

Ivan Smirnov, Leading Research Fellow of the Laboratory of Computational Social Sciences at the Institute of Education of HSE University, has created a computer model that can distinguish high academic achievers from lower ones based on their social media posts. The prediction model uses a mathematical textual analysis that registers users' vocabulary (its range and the semantic fields from which concepts are taken), characters and symbols, post length, and word length. Every word has its own rating (a kind of IQ). Scientific and cultural topics, English words, and words and posts that are longer in length rank highly and serve as indicators of good academic performance. An abundance of emojis, words or whole phrases written in capital letters, and vocabulary related to horoscopes, driving, and military service indicate lower grades in school.

Knowledge tracing is a fundamental task in the computer-aid educational system. In this paper, we propose a hierarchical exercise feature enhanced knowledge tracing framework, which could enhance the ability of knowledge tracing by incorporating knowledge distribution, semantic features, and difficulty features from exercise text. Extensive experiments show the high performance of our framework.

The success of minimax learning problems of generative adversarial networks (GANs) has been observed to depend on the minimax optimization algorithm used for their training. This dependence is commonly attributed to the convergence speed and robustness properties of the underlying optimization algorithm. In this paper, we show that the optimization algorithm also plays a key role in the generalization performance of the trained minimax model. To this end, we analyze the generalization properties of standard gradient descent ascent (GDA) and proximal point method (PPM) algorithms through the lens of algorithmic stability under both convex concave and non-convex non-concave minimax settings. While the GDA algorithm is not guaranteed to have a vanishing excess risk in convex concave problems, we show the PPM algorithm enjoys a bounded excess risk in the same setup. For non-convex non-concave problems, we compare the generalization performance of stochastic GDA and GDmax algorithms where the latter fully solves the maximization subproblem at every iteration. Our generalization analysis suggests the superiority of GDA provided that the minimization and maximization subproblems are solved simultaneously with similar learning rates. We discuss several numerical results indicating the role of optimization algorithms in the generalization of the learned minimax models.

This paper proposes a theoretical analysis of recommendation systems in an online setting, where items are sequentially recommended to users over time. In each round, a user, randomly picked from a population of $m$ users, requests a recommendation. The decision-maker observes the user and selects an item from a catalogue of $n$ items. Importantly, an item cannot be recommended twice to the same user. The probabilities that a user likes each item are unknown. The performance of the recommendation algorithm is captured through its regret, considering as a reference an Oracle algorithm aware of these probabilities. We investigate various structural assumptions on these probabilities: we derive for each structure regret lower bounds, and devise algorithms achieving these limits. Interestingly, our analysis reveals the relative weights of the different components of regret: the component due to the constraint of not presenting the same item twice to the same user, that due to learning the chances users like items, and finally that arising when learning the underlying structure.

Human intelligence is characterized not only by the capacity to learn complex skills, but the ability to rapidly adapt and acquire new skills within an ever-changing environment. In this work we study how the learning of modular solutions can allow for effective generalization to both unseen and potentially differently distributed data. Our main postulate is that the combination of task segmentation, modular learning and memory-based ensembling can give rise to generalization on an exponentially growing number of unseen tasks. We provide a concrete instantiation of this idea using a combination of: (1) the Forget-Me-Not Process, for task segmentation and memory based ensembling; and (2) Gated Linear Networks, which in contrast to contemporary deep learning techniques use a modular and local learning mechanism. We demonstrate that this system exhibits a number of desirable continual learning properties: robustness to catastrophic forgetting, no negative transfer and increasing levels of positive transfer as more tasks are seen. We show competitive performance against both offline and online methods on standard continual learning benchmarks.

Large observational data are increasingly available in disciplines such as health, economic and social sciences, where researchers are interested in causal questions rather than prediction. In this paper, we investigate the problem of estimating heterogeneous treatment effects using non-parametric regression-based methods. Firstly, we introduce the setup and the issues related to conducting causal inference with observational or non-fully randomized data, and how these issues can be tackled with the help of statistical learning tools. Then, we provide a review of state-of-the-art methods, with a particular focus on non-parametric modeling, and we cast them under a unifying taxonomy. After presenting a brief overview on the problem of model selection, we illustrate the performance of some of the methods on three different simulated studies and on a real world example to investigate the effect of participation in school meal programs on health indicators.

Speech synthesis is an important practical generative modeling problem that has seen great progress over the last few years, with likelihood-based autoregressive neural models now outperforming traditional concatenative systems. A downside of such autoregressive models is that they require executing tens of thousands of sequential operations per second of generated audio, making them ill-suited for deployment on specialized deep learning hardware. Here, we propose a new learning method that allows us to train highly parallel models of speech, without requiring access to an analytical likelihood function. Our approach is based on a generalized energy distance between the distributions of the generated and real audio. This spectral energy distance is a proper scoring rule with respect to the distribution over magnitude-spectrograms of the generated waveform audio and offers statistical consistency guarantees. The distance can be calculated from minibatches without bias, and does not involve adversarial learning, yielding a stable and consistent method for training implicit generative models. Empirically, we achieve state-of-the-art generation quality among implicit generative models, as judged by the recently-proposed cFDSD metric. When combining our method with adversarial techniques, we also improve upon the recently-proposed GAN-TTS model in terms of Mean Opinion Score as judged by trained human evaluators.

Machine learning (ML) tasks are becoming ubiquitous in today's network applications. Federated learning has emerged recently as a technique for training ML models at the network edge by leveraging processing capabilities across the nodes that collect the data. There are several challenges with employing conventional federated learning in contemporary networks, due to the significant heterogeneity in compute and communication capabilities that exist across devices. To address this, we advocate a new learning paradigm called fog learning which will intelligently distribute ML model training across the continuum of nodes from edge devices to cloud servers. Fog learning enhances federated learning along three major dimensions: network, heterogeneity, and proximity. It considers a multi-layer hybrid learning framework consisting of heterogeneous devices with various proximities. It accounts for the topology structures of the local networks among the heterogeneous nodes at each network layer, orchestrating them for collaborative/cooperative learning through device-to-device (D2D) communications. This migrates from star network topologies used for parameter transfers in federated learning to more distributed topologies at scale. We discuss several open research directions to realizing fog learning.