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.

Anti-Exploration by Random Network Distillation

Despite the success of Random Network Distillation (RND) in various domains, it was shown as not discriminative enough to be used as an uncertainty estimator for penalizing out-of-distribution actions in offline reinforcement learning. In this paper, we revisit these results and show that, with a naive choice of conditioning for the RND prior, it becomes infeasible for the actor to effectively minimize the anti-exploration bonus and discriminativity is not an issue. We show that this limitation can be avoided with conditioning based on Feature-wise Linear Modulation (FiLM), resulting in a simple and efficient ensemble-free algorithm based on Soft Actor-Critic. We evaluate it on the D4RL benchmark, showing that it is capable of achieving performance comparable to ensemble-based methods and outperforming ensemble-free approaches by a wide margin.

Q-Ensemble for Offline RL: Don't Scale the Ensemble, Scale the Batch Size

Training large neural networks is known to be time-consuming, with the learning duration taking days or even weeks. To address this problem, large-batch optimization was introduced. This approach demonstrated that scaling mini-batch sizes with appropriate learning rate adjustments can speed up the training process by orders of magnitude. While long training time was not typically a major issue for model-free deep offline RL algorithms, recently introduced Q-ensemble methods achieving state-of-the-art performance made this issue more relevant, notably extending the training duration. In this work, we demonstrate how this class of methods can benefit from large-batch optimization, which is commonly overlooked by the deep offline RL community. We show that scaling the mini-batch size and naively adjusting the learning rate allows for (1) a reduced size of the Q-ensemble, (2) stronger penalization of out-of-distribution actions, and (3) improved convergence time, effectively shortening training duration by 3-4x times on average.

Let Offline RL Flow: Training Conservative Agents in the Latent Space of Normalizing Flows

Offline reinforcement learning aims to train a policy on a pre-recorded and fixed dataset without any additional environment interactions. There are two major challenges in this setting: (1) extrapolation error caused by approximating the value of state-action pairs not well-covered by the training data and (2) distributional shift between behavior and inference policies. One way to tackle these problems is to induce conservatism - i.e., keeping the learned policies closer to the behavioral ones. To achieve this, we build upon recent works on learning policies in latent action spaces and use a special form of Normalizing Flows for constructing a generative model, which we use as a conservative action encoder. This Normalizing Flows action encoder is pre-trained in a supervised manner on the offline dataset, and then an additional policy model - controller in the latent space - is trained via reinforcement learning. This approach avoids querying actions outside of the training dataset and therefore does not require additional regularization for out-of-dataset actions. We evaluate our method on various locomotion and navigation tasks, demonstrating that our approach outperforms recently proposed algorithms with generative action models on a large portion of datasets.

CORL: Research-oriented Deep Offline Reinforcement Learning Library

CORL is an open-source library that provides single-file implementations of Deep Offline Reinforcement Learning algorithms. It emphasizes a simple developing experience with a straightforward codebase and a modern analysis tracking tool. In CORL, we isolate methods implementation into distinct single files, making performance-relevant details easier to recognise. Additionally, an experiment tracking feature is available to help log metrics, hyperparameters, dependencies, and more to the cloud. Finally, we have ensured the reliability of the implementations by benchmarking a commonly employed D4RL benchmark. The source code can be found https://github.com/tinkoff-ai/CORL

Deep Image Retrieval is not Robust to Label Noise

Large-scale datasets are essential for the success of deep learning in image retrieval. However, manual assessment errors and semi-supervised annotation techniques can lead to label noise even in popular datasets. As previous works primarily studied annotation quality in image classification tasks, it is still unclear how label noise affects deep learning approaches to image retrieval. In this work, we show that image retrieval methods are less robust to label noise than image classification ones. Furthermore, we, for the first time, investigate different types of label noise specific to image retrieval tasks and study their effect on model performance.

EXACT: How to Train Your Accuracy

Classification tasks are usually evaluated in terms of accuracy. However, accuracy is discontinuous and cannot be directly optimized using gradient ascent. Popular methods minimize cross-entropy, Hinge loss, or other surrogate losses, which can lead to suboptimal results. In this paper, we propose a new optimization framework by introducing stochasticity to a model's output and optimizing expected accuracy, i.e. accuracy of the stochastic model. Extensive experiments on image classification show that the proposed optimization method is a powerful alternative to widely used classification losses.

CVTT: Cross-Validation Through Time

The practical aspects of evaluating recommender systems is an actively discussed topic in the research community. While many current evaluation techniques bring performance down to a single-value metric as a straightforward approach for model comparison, it is based on a strong assumption of the methods' stable performance over time. In this paper, we argue that leaving out a method's continuous performance can lead to losing valuable insight into joint data-method effects. We propose the Cross-Validation Thought Time (CVTT) technique to perform more detailed evaluations, which focus on model cross-validation performance over time. Using the proposed technique, we conduct a detailed analysis of popular RecSys algorithms' performance against various metrics and datasets. We also compare several data preparation and evaluation strategies to analyze their impact on model performance. Our results show that model performance can vary significantly over time, and both data and evaluation setup can have a marked effect on it.

Probabilistic Embeddings Revisited

In recent years, deep metric learning and its probabilistic extensions achieved state-of-the-art results in a face verification task. However, despite improvements in face verification, probabilistic methods received little attention in the community. It is still unclear whether they can improve image retrieval quality. In this paper, we present an extensive comparison of probabilistic methods in verification and retrieval tasks. Following the suggested methodology, we outperform metric learning baselines using probabilistic methods and propose several directions for future work and improvements.