Bayesian optimization is an effective method for finding extrema of a black-box function. We propose a new type of Bayesian optimization for learning user preferences in high-dimensional spaces. The central assumption is that the underlying objective function cannot be evaluated directly, but instead a minimizer along a projection can be queried, which we call a projective preferential query. The form of the query allows for feedback that is natural for a human to give, and which enables interaction. This is demonstrated in a user experiment in which the user feedback comes in the form of optimal position and orientation of a molecule adsorbing to a surface. We demonstrate that our framework is able to find a global minimum of a high-dimensional black-box function, which is an infeasible task for existing preferential Bayesian optimization frameworks that are based on pairwise comparisons.

Conferences and vendor marketing materials are full of trite and banal sayings. Say something that seems to be profound, and perhaps they'll think that everything else you have to say is just as profound. One of the common refrains you might hear at many an AI and data-focused event is the pithy statement that "data is the new oil" as if that's supposed to mean something profound. The first time I heard this expression (about a decade ago, I should add), it was an interesting point to make about how "important" and "strategic" data is. But every time I've heard it since, it's bandied about to imply something more than it is.

As the nations of Earth attempt to invent and implement solutions to the growing threat of climate change, just about every option is on the table. Investing in renewable sources of energy and dropping emissions around the globe are the dominant strategies, but utilizing artificial intelligence can help reduce the damage done by climate change. As reported by Live Mint, artificial intelligence algorithms can help conservationists limit deforestation, protect vulnerable species of animals from climate change, fight poaching, and monitor air pollution. The data science company Gramener has employed machine learning to help get estimates of the number of penguin colonies across Antarctica by analyzing images taken by camera traps. The size of penguin colonies in Antarctica has decreased dramatically over the course of the past decade, impacted by climate change.

Training machine learning models that can learn complex spatiotemporal dynamics and generalize under distributional shift is a fundamental challenge. The symmetries in a physical system play a unique role in characterizing unchanged features under transformation. We propose a systematic approach to improve generalization in spatiotemporal models by incorporating symmetries into deep neural networks. Our general framework to design equivariant convolutional models employs (1) convolution with equivariant kernels, (2) conjugation by averaging operators in order to force equivariance, (3) and a naturally equivariant generalization of convolution called group correlation. Our framework is both theoretically and experimentally robust to distributional shift by a symmetry group and enjoys favorable sample complexity. We demonstrate the advantage of our approach on a variety of physical dynamics including turbulence and diffusion systems. This is the first time that equivariant CNNs have been used to forecast physical dynamics.

This paper presents a distributionally robust Q-Learning algorithm (DrQ) which leverages Wasserstein ambiguity sets to provide probabilistic out-of-sample safety guarantees during online learning. First, we follow past work by separating the constraint functions from the principal objective to create a hierarchy of machines within the constrained Markov decision process (CMDP). DrQ works within this framework by augmenting constraint costs with tightening offset variables obtained through Wasserstein distributionally robust optimization (DRO). These offset variables correspond to worst-case distributions of modeling error characterized by the TD-errors of the constraint Q-functions. This overall procedure allows us to safely approach the nominal constraint boundaries with strong probabilistic out-of-sample safety guarantees. Using a case study of safe lithium-ion battery fast charging, we demonstrate dramatic improvements in safety and performance relative to a conventional DQN.

There is broad interest in creating RL agents that can solve many (related) tasks and adapt to new tasks and environments after initial training. Model-based RL leverages learned surrogate models that describe dynamics and rewards of individual tasks, such that planning in a good surrogate can lead to good control of the true system. Rather than solving each task individually from scratch, hierarchical models can exploit the fact that tasks are often related by (unobserved) causal factors of variation in order to achieve efficient generalization, as in learning how the mass of an item affects the force required to lift it can generalize to previously unobserved masses. We propose Generalized Hidden Parameter MDPs (GHP-MDPs) that describe a family of MDPs where both dynamics and reward can change as a function of hidden parameters that vary across tasks. The GHP-MDP augments model-based RL with latent variables that capture these hidden parameters, facilitating transfer across tasks. We also explore a variant of the model that incorporates explicit latent structure mirroring the causal factors of variation across tasks (for instance: agent properties, environmental factors, and goals). We experimentally demonstrate state-of-the-art performance and sample-efficiency on a new challenging MuJoCo task using reward and dynamics latent spaces, while beating a previous state-of-the-art baseline with $>10\times$ less data. Using test-time inference of the latent variables, our approach generalizes in a single episode to novel combinations of dynamics and reward, and to novel rewards.

Artificial intelligence is at the peak of its hype curve, and its applications in the solar energy sector are amid a surge in popularity. Once upon a time confined solely to the domains of science fiction, this technology is transforming the energy landscape, altering how solar assets are managed, operated, and maintained. Year after year, the cumulative global PV capacity is increasing by gigawatts, which are highly dependent on operating conditions that are inherently variable and hard to predict. Also, further consolidation of these solar assets is leading to these portfolios growing not only in size but also in dispersity. These factors have made managing solar assets considerably more challenging.

A team of Fugro employees has won a global competition in geotechnical machine-learning. Competing with 60 other teams from industry and academia around the world, the Fugro team came first in the pile-driving prediction event organised as part of the International Symposium on Frontiers in Offshore Geotechnics (ISFOG) 2020 conference, which will be held in Austin, Texas, in August. The competition ran from April to December 2019, and ended on 1 January 2020, when it was announced that Fugro had won.

Globally, thematic investing has tripled over the past five years to around $40.76 billion, per Morningstar Inc. This is steadily taking over the investment world, largely due to the introduction of theme-based funds and also for its long-term and easy-to-comprehend approach. Thematic investing requires investment in companies that can benefit from the technological, demographic and environmental changes (read: Top ETF Areas for 2020). Let's take a look at some of the themes that are currently in vogue. We are living in an era that is largely dominated by AI applications and technological advancements.

This paper focuses on deep reinforcement learning (DRL)-based energy dispatch for isolated microgrids (MGs) with diesel generators (DGs), photovoltaic (PV) panels, and a battery. A finite-horizon Partial Observable Markov Decision Process (POMDP) model is formulated and solved by learning from historical data to capture the uncertainty in future electricity consumption and renewable power generation. In order to deal with the instability problem of DRL algorithms and unique characteristics of finite-horizon models, two novel DRL algorithms, namely, FH-DDPG and FH-RDPG, are proposed to derive energy dispatch policies with and without fully observable state information. A case study using real isolated microgrid data is performed, where the performance of the proposed algorithms are compared with the myopic algorithm as well as other baseline DRL algorithms. Moreover, the impact of uncertainties on MG performance is decoupled into two levels and evaluated respectively.