Gradient-based meta-learners such as Model-Agnostic Meta-Learning (MAML) have shown strong few-shot performance in supervised and reinforcement learning settings. However, specifically in the case of meta-reinforcement learning (meta-RL), we can show that gradient-based meta-learners are sensitive to task distributions. With the wrong curriculum, agents suffer the effects of meta-overfitting, shallow adaptation, and adaptation instability. In this work, we begin by highlighting intriguing failure cases of gradient-based meta-RL and show that task distributions can wildly affect algorithmic outputs, stability, and performance. To address this problem, we leverage insights from recent literature on domain randomization and propose meta Active Domain Randomization (meta-ADR), which learns a curriculum of tasks for gradient-based meta-RL in a similar as ADR does for sim2real transfer. We show that this approach induces more stable policies on a variety of simulated locomotion and navigation tasks. We assess in- and out-of-distribution generalization and find that the learned task distributions, even in an unstructured task space, greatly improve the adaptation performance of MAML. Finally, we motivate the need for better benchmarking in meta-RL that prioritizes \textit{generalization} over single-task adaption performance.

Goal-directed Reinforcement Learning (RL) traditionally considers an agent interacting with an environment, prescribing a real-valued reward to an agent proportional to the completion of some goal. Goal-directed RL has seen large gains in sample efficiency, due to the ease of reusing or generating new experience by proposing goals. In this work, we build on the framework of self-play, allowing an agent to interact with itself in order to make progress on some unknown task. We use Active Domain Randomization and self-play to create a novel, coupled environment-goal curriculum, where agents learn through progressively more difficult tasks and environment variations. Our method, Self-Supervised Active Domain Randomization (SS-ADR), generates a growing curriculum, encouraging the agent to try tasks that are just outside of its current capabilities, while building a domain-randomization curriculum that enables state-of-the-art results on various sim2real transfer tasks. Our results show that a curriculum of co-evolving the environment difficulty along with the difficulty of goals set in each environment provides practical benefits in the goal-directed tasks tested.

Virginia Commonwealth University School of the Arts in Qatar (VCUarts Qatar) is hosting a lecture by Dr James She from Hong Kong University of Science and Technology (HKUST), titled, 'AI and Smartphone Technologies for New Artwork Creation, Interaction and Definition'. The lecture will take place this Tuesday, 18 February, 12:30 pm, at the VCUarts Qatar Atrium. The event is open to everyone. Emerging Artificial Intelligence (AI) and smartphone technologies are making disruptive changes and new possibilities for people in business, manufacturing, travel, education and even art. In this talk, examples of AI and smartphone technologies will be selected to show how recent developments in those technologies could facilitate the creation of, and interaction with, artworks.

Artificial intelligence--code that learns--is likely to be humankind's most important invention. It's a 60-year-old idea that took off five years ago when fast chips enabled massive computing and sensors, cameras, and robots fed data-hungry algorithms. We're a couple of years into a new age where machine learning (a functional subset of AI), big data and enabling technologies are transforming every sector. In every sector, there is a big data set behind every question. Every field is computational: healthcare, manufacturing, law, finance and accounting, retail, and real estate.

In Washington, DC, where big decisions are made daily, an assembly of executives, data scientists and analytics professionals will gather for the world's premier analytics conference, SAS Global Forum. During the event, March 29 through April 1, these analytics enthusiasts will share insights and learn how analytics and artificial intelligence (AI) help businesses make better, faster decisions that improve lives around the globe. This year's conference will transform the Walter E. Washington Convention Center into an interactive analytics showcase. Featuring more than 600 sessions from leading organizations across industries, SAS Global Forum also includes hands-on workshops, e-learning, training, demos and networking opportunities, all designed to empower users and executives to "Do Great Things," this year's theme. Now in its 44th year, SAS Global Forum showcases the technologies and industry topics that analytics professionals most want on the agenda.

We're experiencing a profound shift in how educational content is created and delivered. Online education platforms are flourishing, and a recent Pew Research poll shows the world's leading video platform YouTube has finally become more than a repository of cat videos -- about half of YouTube users say it's very important for helping them figure out how to do things they've never done before. The year 2019 saw unprecedented growth in YouTube educational content on artificial intelligence. Synced has selected 10 AI-oriented YouTube channels we hope might provide our readers a cozy little holiday binge-watching session. Preserve Knowledge focuses on advances in mathematics, computer science, and artificial intelligence.

We consider the task of automated theorem proving, a key AI task. Deep learning has shown promise for training theorem provers, but there are limited human-written theorems and proofs available for supervised learning. To address this limitation, we propose to learn a neural generator that automatically synthesizes theorems and proofs for the purpose of training a theorem prover. Experiments on real-world tasks demonstrate that synthetic data from our approach improves the theorem prover and advances the state of the art of automated theorem proving in Metamath.

We develop two new stochastic Gauss-Newton algorithms for solving a class of stochastic nonconvex compositional optimization problems frequently arising in practice. We consider both the expectation and finite-sum settings under standard assumptions. We use both classical stochastic and SARAH estimators for approximating function values and Jacobians. In the expectation case, we establish $\mathcal{O}(\varepsilon^{-2})$ iteration complexity to achieve a stationary point in expectation and estimate the total number of stochastic oracle calls for both function values and its Jacobian, where $\varepsilon$ is a desired accuracy. In the finite sum case, we also estimate the same iteration complexity and the total oracle calls with high probability. To our best knowledge, this is the first time such global stochastic oracle complexity is established for stochastic Gauss-Newton methods. We illustrate our theoretical results via numerical examples on both synthetic and real datasets.

As algorithmic decision-making takes a more and more important role in myriad application domains, incentives emerge to change the inputs presented to these algorithms--people may either invest in truly relevant attributes or strategically lie about their data. Recently, a collection of very interesting papers has explored various models of strategic behavior on the part of the classified individuals in learning settings, and ways to mitigate the harms to accuracy that can arise from falsified features [Dalvi et al., 2004, Brückner et al., 2012, Hardt et al., 2016, Dong et al., 2018]. Additionally, some recent work has focused on the design of learning algorithms that incentivize the classified individuals to make"good" investments in true changes to their variables[Kleinberg and Raghavan, 2019]. The present paper takes a different tack, and explores another potential effect of strategic investment in true changes to variables, in an online learning setting: we claim that interaction between the online learning and the strategic individuals may actually aid the learning algorithm in identifying causal variables. By causal, we mean, informally, variables such that changes in their true value cause changes in the true label and lead agents to improve.

We demonstrate how to scalably solve a class of constrained self-concordant minimization problems using linear minimization oracles (LMO) over the constraint set. We prove that the number of LMO calls of our method is nearly the same as that of the Frank-Wolfe method in L-smooth case. Specifically, our Newton Frank-Wolfe method uses O (ε ν) LMO's, where ε is the desired accuracy and ν: 1 o(1). In addition, we demonstrate how our algorithm can exploit the improved variants of the LMO-based schemes, including away-steps, to attain linear convergence rates. We also provide numerical evidence with portfolio design with the competitive ratio, D-optimal experimental design, and logistic regression with the elastic net where Newton Frank-Wolfe outperforms the state-of-the-art.