Search for: All records

Apprenticeship learning crucially depends on effectively learning rewards, and hence control policies from user demonstrations. Of particular difficulty is the setting where the desired task consists of a number of sub-goals with temporal dependencies. The quality of inferred rewards and hence policies are typically limited by the quality of demonstrations, and poor inference of these can lead to undesirable outcomes. In this paper, we show how temporal logic specifications that describe high level task objectives, are encoded in a graph to define a temporal-based metric that reasons about behaviors of demonstrators and the learner agent to improve the quality of inferred rewards and policies. Through experiments on a diverse set of robot manipulator simulations, we show how our framework overcomes the drawbacks of prior literature by drastically improving the number of demonstrations required to learn a control policy. 

Researchers in human–robot collaboration have extensively studied methods for inferring human intentions and predicting their actions, as this is an important precursor for robots to provide useful assistance. We review contemporary methods for intention inference and human activity prediction. Our survey finds that intentions and goals are often inferred via Bayesian posterior estimation and Markov decision processes that model internal human states as unobserved variables or represent both agents in a shared probabilistic framework. An alternative approach is to use neural networks and other supervised learning approaches to directly map observable outcomes to intentions and to make predictions about future human activity based on past observations. That said, due to the complexity of human intentions, existing work usually reasons about limited domains, makes unrealistic simplifications about intentions, and is mostly constrained to short-term predictions. This state of the art provides opportunity for future research that could include more nuanced models of intents, reason over longer horizons, and account for the human tendency to adapt. 

Single-objective optimization algorithms search for the single highest quality solution with respect to an objective. Quality diversity (QD) optimization algorithms, such as Covariance Matrix Adaptation MAP-Elites (CMA-ME), search for a collection of solutions that are both high quality with respect to an objective and diverse with respect to specified measure functions. However, CMA-ME suffers from three major limitations highlighted by the QD community: prematurely abandoning the objective in favor of exploration, struggling to explore flat objectives, and having poor performance for low-resolution archives. We propose a new QD algorithm, CMA MAP-Annealing (CMA-MAE), and its differentiable QD variant, CMA-MAE via a Gradient Arborescence (CMA-MAEGA), that address all three limitations. We provide theoretical justifications for the new algorithm with respect to each limitation. Our theory informs our experiments, which support the theory and show that CMA-MAE achieves state-of-the-art performance and robustness on standard QD benchmark and reinforcement learning domains.