In-Context Reinforcement Learning for Variable Action Spaces

Recent work has shown that supervised pre-training on learning histories of RL algorithms results in a model that captures the learning process and is able to improve in-context on novel tasks through interactions with an environment. Despite the progress in this area, there is still a gap in the existing literature, particularly in the in-context generalization to new action spaces. While existing methods show high performance on new tasks created by different reward distributions, their architectural design and training process are not suited for the introduction of new actions during evaluation. We aim to bridge this gap by developing an architecture and training methodology specifically for the task of generalizing to new action spaces. Inspired by Headless LLM, we remove the dependence on the number of actions by directly predicting the action embeddings. Furthermore, we use random embeddings to force the semantic inference of actions from context and to prepare for the new unseen embeddings during test time. Using multi-armed bandit environments with a variable number of arms, we show that our model achieves the performance of the data generation algorithm without requiring retraining for each new environment.

Emergence of In-Context Reinforcement Learning from Noise Distillation

In-Context Reinforcement Learning is an emerging field with great potential for advancing Artificial Intelligence. Its core capability lies in generalizing to unseen tasks through interaction with the environment. To master these capabilities, an agent must be trained on specifically curated data that includes a policy improvement that an algorithm seeks to extract and then apply in context in the environment. However, for numerous tasks, training RL agents may be unfeasible, while obtaining human demonstrations can be relatively easy. Additionally, it is rare to be given the optimal policy, typically, only suboptimal demonstrations are available. We propose $AD^{\epsilon}$, a method that leverages demonstrations without policy improvement and enables multi-task in-context learning in the presence of a suboptimal demonstrator. This is achieved by artificially creating a history of incremental improvement, wherein noise is systematically introduced into the demonstrator's policy. Consequently, each successive transition illustrates a marginally better trajectory than the previous one. Our approach was tested on the Dark Room and Dark Key-to-Door environments, resulting in over a $\textbf{2}$x improvement compared to the best available policy in the data.

XLand-MiniGrid: Scalable Meta-Reinforcement Learning Environments in JAX

We present XLand-MiniGrid, a suite of tools and grid-world environments for meta-reinforcement learning research inspired by the diversity and depth of XLand and the simplicity and minimalism of MiniGrid. XLand-Minigrid is written in JAX, designed to be highly scalable, and can potentially run on GPU or TPU accelerators, democratizing large-scale experimentation with limited resources. To demonstrate the generality of our library, we have implemented some well-known single-task environments as well as new meta-learning environments capable of generating $10^8$ distinct tasks. We have empirically shown that the proposed environments can scale up to $2^{13}$ parallel instances on the GPU, reaching tens of millions of steps per second.

Unveiling Empirical Pathologies of Laplace Approximation for Uncertainty Estimation

In this paper, we critically evaluate Bayesian methods for uncertainty estimation in deep learning, focusing on the widely applied Laplace approximation and its variants. Our findings reveal that the conventional method of fitting the Hessian matrix negatively impacts out-of-distribution (OOD) detection efficiency. We propose a different point of view, asserting that focusing solely on optimizing prior precision can yield more accurate uncertainty estimates in OOD detection while preserving adequate calibration metrics. Moreover, we demonstrate that this property is not connected to the training stage of a model but rather to its intrinsic properties. Through extensive experimental evaluation, we establish the superiority of our simplified approach over traditional methods in the out-of-distribution domain.

Wild-Tab: A Benchmark For Out-Of-Distribution Generalization In Tabular Regression

Out-of-Distribution (OOD) generalization, a cornerstone for building robust machine learning models capable of handling data diverging from the training set's distribution, is an ongoing challenge in deep learning. While significant progress has been observed in computer vision and natural language processing, its exploration in tabular data, ubiquitous in many industrial applications, remains nascent. To bridge this gap, we present Wild-Tab, a large-scale benchmark tailored for OOD generalization in tabular regression tasks. The benchmark incorporates 3 industrial datasets sourced from fields like weather prediction and power consumption estimation, providing a challenging testbed for evaluating OOD performance under real-world conditions. Our extensive experiments, evaluating 10 distinct OOD generalization methods on Wild-Tab, reveal nuanced insights. We observe that many of these methods often struggle to maintain high-performance levels on unseen data, with OOD performance showing a marked drop compared to in-distribution performance. At the same time, Empirical Risk Minimization (ERM), despite its simplicity, delivers robust performance across all evaluations, rivaling the results of state-of-the-art methods. Looking forward, we hope that the release of Wild-Tab will facilitate further research on OOD generalization and aid in the deployment of machine learning models in various real-world contexts where handling distribution shifts is a crucial requirement.

Time-Aware Item Weighting for the Next Basket Recommendations

In this paper we study the next basket recommendation problem. Recent methods use different approaches to achieve better performance. However, many of them do not use information about the time of prediction and time intervals between baskets. To fill this gap, we propose a novel method, Time-Aware Item-based Weighting (TAIW), which takes timestamps and intervals into account. We provide experiments on three real-world datasets, and TAIW outperforms well-tuned state-of-the-art baselines for next-basket recommendations. In addition, we show the results of an ablation study and a case study of a few items.

RecBaselines2023: a new dataset for choosing baselines for recommender models

The number of proposed recommender algorithms continues to grow. The authors propose new approaches and compare them with existing models, called baselines. Due to the large number of recommender models, it is difficult to estimate which algorithms to choose in the article. To solve this problem, we have collected and published a dataset containing information about the recommender models used in 903 papers, both as baselines and as proposed approaches. This dataset can be seen as a typical dataset with interactions between papers and previously proposed models. In addition, we provide a descriptive analysis of the dataset and highlight possible challenges to be investigated with the data. Furthermore, we have conducted extensive experiments using a well-established methodology to build a good recommender algorithm under the dataset. Our experiments show that the selection of the best baselines for proposing new recommender approaches can be considered and successfully solved by existing state-of-the-art collaborative filtering models. Finally, we discuss limitations and future work.

Katakomba: Tools and Benchmarks for Data-Driven NetHack

NetHack is known as the frontier of reinforcement learning research where learning-based methods still need to catch up to rule-based solutions. One of the promising directions for a breakthrough is using pre-collected datasets similar to recent developments in robotics, recommender systems, and more under the umbrella of offline reinforcement learning (ORL). Recently, a large-scale NetHack dataset was released; while it was a necessary step forward, it has yet to gain wide adoption in the ORL community. In this work, we argue that there are three major obstacles for adoption: tool-wise, implementation-wise, and benchmark-wise. To address them, we develop an open-source library that provides workflow fundamentals familiar to the ORL community: pre-defined D4RL-style tasks, uncluttered baseline implementations, and reliable evaluation tools with accompanying configs and logs synced to the cloud.

Diversifying Deep Ensembles: A Saliency Map Approach for Enhanced OOD Detection, Calibration, and Accuracy

Deep ensembles achieved state-of-the-art results in classification and out-of-distribution (OOD) detection; however, their effectiveness remains limited due to the homogeneity of learned patterns within the ensemble. To overcome this challenge, our study introduces a novel approach that promotes diversity among ensemble members by leveraging saliency maps. By incorporating saliency map diversification, our method outperforms conventional ensemble techniques in multiple classification and OOD detection tasks, while also improving calibration. Experiments on well-established OpenOOD benchmarks highlight the potential of our method in practical applications.

Revisiting the Minimalist Approach to Offline Reinforcement Learning

Recent years have witnessed significant advancements in offline reinforcement learning (RL), resulting in the development of numerous algorithms with varying degrees of complexity. While these algorithms have led to noteworthy improvements, many incorporate seemingly minor design choices that impact their effectiveness beyond core algorithmic advances. However, the effect of these design choices on established baselines remains understudied. In this work, we aim to bridge this gap by conducting a retrospective analysis of recent works in offline RL and propose ReBRAC, a minimalistic algorithm that integrates such design elements built on top of the TD3+BC method. We evaluate ReBRAC on 51 datasets with both proprioceptive and visual state spaces using D4RL and V-D4RL benchmarks, demonstrating its state-of-the-art performance among ensemble-free methods. To further illustrate the efficacy of these design choices, we perform a large-scale ablation study and hyperparameter sensitivity analysis on the scale of thousands of experiments.