Lili Wu

Ian Waudby-Smith, Lili Wu, Aaditya Ramdas et al.

Contextual bandit algorithms are ubiquitous tools for active sequential experimentation in healthcare and the tech industry. They involve online learning algorithms that adaptively learn policies over time to map observed contexts $X_t$ to actions $A_t$ in an attempt to maximize stochastic rewards $R_t$. This adaptivity raises interesting but hard statistical inference questions, especially counterfactual ones: for example, it is often of interest to estimate the properties of a hypothetical policy that is different from the logging policy that was used to collect the data -- a problem known as ``off-policy evaluation'' (OPE). Using modern martingale techniques, we present a comprehensive framework for OPE inference that relax unnecessary conditions made in some past works, significantly improving on them both theoretically and empirically. Importantly, our methods can be employed while the original experiment is still running (that is, not necessarily post-hoc), when the logging policy may be itself changing (due to learning), and even if the context distributions are a highly dependent time-series (such as if they are drifting over time). More concretely, we derive confidence sequences for various functionals of interest in OPE. These include doubly robust ones for time-varying off-policy mean reward values, but also confidence bands for the entire cumulative distribution function of the off-policy reward distribution. All of our methods (a) are valid at arbitrary stopping times (b) only make nonparametric assumptions, (c) do not require importance weights to be uniformly bounded and if they are, we do not need to know these bounds, and (d) adapt to the empirical variance of our estimators. In summary, our methods enable anytime-valid off-policy inference using adaptively collected contextual bandit data.

Anurag Koul, Shivakanth Sujit, Shaoru Chen et al.

Goal-conditioned planning benefits from learned low-dimensional representations of rich observations. While compact latent representations typically learned from variational autoencoders or inverse dynamics enable goal-conditioned decision making, they ignore state reachability, hampering their performance. In this paper, we learn a representation that associates reachable states together for effective planning and goal-conditioned policy learning. We first learn a latent representation with multi-step inverse dynamics (to remove distracting information), and then transform this representation to associate reachable states together in $\ell_2$ space. Our proposals are rigorously tested in various simulation testbeds. Numerical results in reward-based settings show significant improvements in sampling efficiency. Further, in reward-free settings this approach yields layered state abstractions that enable computationally efficient hierarchical planning for reaching ad hoc goals with zero additional samples.

Lili Wu, Ben Evans, Riashat Islam et al.

Discovering an informative, or agent-centric, state representation that encodes only the relevant information while discarding the irrelevant is a key challenge towards scaling reinforcement learning algorithms and efficiently applying them to downstream tasks. Prior works studied this problem in high-dimensional Markovian environments, when the current observation may be a complex object but is sufficient to decode the informative state. In this work, we consider the problem of discovering the agent-centric state in the more challenging high-dimensional non-Markovian setting, when the state can be decoded from a sequence of past observations. We establish that generalized inverse models can be adapted for learning agent-centric state representation for this task. Our results include asymptotic theory in the deterministic dynamics setting as well as counter-examples for alternative intuitive algorithms. We complement these findings with a thorough empirical study on the agent-centric state discovery abilities of the different alternatives we put forward. Particularly notable is our analysis of past actions, where we show that these can be a double-edged sword: making the algorithms more successful when used correctly and causing dramatic failure when used incorrectly.

Jesse Clifton, Lili Wu, Eric Laber

We introduce Parameterized Exploration (PE), a simple family of methods for model-based tuning of the exploration schedule in sequential decision problems. Unlike common heuristics for exploration, our method accounts for the time horizon of the decision problem as well as the agent's current state of knowledge of the dynamics of the decision problem. We show our method as applied to several common exploration techniques has superior performance relative to un-tuned counterparts in Bernoulli and Gaussian multi-armed bandits, contextual bandits, and a Markov decision process based on a mobile health (mHealth) study. We also examine the effects of the accuracy of the estimated dynamics model on the performance of PE.