RAISE: Enhancing Scientific Reasoning in LLMs via Step-by-Step Retrieval

Scientific reasoning requires not only long-chain reasoning processes, but also knowledge of domain-specific terminologies and adaptation to updated findings. To deal with these challenges for scientific reasoning, we introduce RAISE, a step-by-step retrieval-augmented framework which retrieves logically relevant documents from in-the-wild corpus. RAISE is divided into three steps: problem decomposition, logical query generation, and logical retrieval. We observe that RAISE consistently outperforms other baselines on scientific reasoning benchmarks. We analyze that unlike other baselines, RAISE retrieves documents that are not only similar in terms of the domain knowledge, but also documents logically more relevant.

Policy-labeled Preference Learning: Is Preference Enough for RLHF?

To design rewards that align with human goals, Reinforcement Learning from Human Feedback (RLHF) has emerged as a prominent technique for learning reward functions from human preferences and optimizing policies via reinforcement learning algorithms. However, existing RLHF methods often misinterpret trajectories as being generated by an optimal policy, causing inaccurate likelihood estimation and suboptimal learning. Inspired by Direct Preference Optimization framework which directly learns optimal policy without explicit reward, we propose policy-labeled preference learning (PPL), to resolve likelihood mismatch issues by modeling human preferences with regret, which reflects behavior policy information. We also provide a contrastive KL regularization, derived from regret-based principles, to enhance RLHF in sequential decision making. Experiments in high-dimensional continuous control tasks demonstrate PPL's significant improvements in offline RLHF performance and its effectiveness in online settings.

Constructing Fair Latent Space for Intersection of Fairness and Explainability

As the use of machine learning models has increased, numerous studies have aimed to enhance fairness. However, research on the intersection of fairness and explainability remains insufficient, leading to potential issues in gaining the trust of actual users. Here, we propose a novel module that constructs a fair latent space, enabling faithful explanation while ensuring fairness. The fair latent space is constructed by disentangling and redistributing labels and sensitive attributes, allowing the generation of counterfactual explanations for each type of information. Our module is attached to a pretrained generative model, transforming its biased latent space into a fair latent space. Additionally, since only the module needs to be trained, there are advantages in terms of time and cost savings, without the need to train the entire generative model. We validate the fair latent space with various fairness metrics and demonstrate that our approach can effectively provide explanations for biased decisions and assurances of fairness.

Mix from Failure: Confusion-Pairing Mixup for Long-Tailed Recognition

Long-tailed image recognition is a computer vision problem considering a real-world class distribution rather than an artificial uniform. Existing methods typically detour the problem by i) adjusting a loss function, ii) decoupling classifier learning, or iii) proposing a new multi-head architecture called experts. In this paper, we tackle the problem from a different perspective to augment a training dataset to enhance the sample diversity of minority classes. Specifically, our method, namely Confusion-Pairing Mixup (CP-Mix), estimates the confusion distribution of the model and handles the data deficiency problem by augmenting samples from confusion pairs in real-time. In this way, CP-Mix trains the model to mitigate its weakness and distinguish a pair of classes it frequently misclassifies. In addition, CP-Mix utilizes a novel mixup formulation to handle the bias in decision boundaries that originated from the imbalanced dataset. Extensive experiments demonstrate that CP-Mix outperforms existing methods for long-tailed image recognition and successfully relieves the confusion of the classifier.

Mitigating Spurious Correlations via Disagreement Probability

Models trained with empirical risk minimization (ERM) are prone to be biased towards spurious correlations between target labels and bias attributes, which leads to poor performance on data groups lacking spurious correlations. It is particularly challenging to address this problem when access to bias labels is not permitted. To mitigate the effect of spurious correlations without bias labels, we first introduce a novel training objective designed to robustly enhance model performance across all data samples, irrespective of the presence of spurious correlations. From this objective, we then derive a debiasing method, Disagreement Probability based Resampling for debiasing (DPR), which does not require bias labels. DPR leverages the disagreement between the target label and the prediction of a biased model to identify bias-conflicting samples-those without spurious correlations-and upsamples them according to the disagreement probability. Empirical evaluations on multiple benchmarks demonstrate that DPR achieves state-of-the-art performance over existing baselines that do not use bias labels. Furthermore, we provide a theoretical analysis that details how DPR reduces dependency on spurious correlations.

Tractable and Provably Efficient Distributional Reinforcement Learning with General Value Function Approximation

Distributional reinforcement learning improves performance by effectively capturing environmental stochasticity, but a comprehensive theoretical understanding of its effectiveness remains elusive. In this paper, we present a regret analysis for distributional reinforcement learning with general value function approximation in a finite episodic Markov decision process setting. We first introduce a key notion of Bellman unbiasedness for a tractable and exactly learnable update via statistical functional dynamic programming. Our theoretical results show that approximating the infinite-dimensional return distribution with a finite number of moment functionals is the only method to learn the statistical information unbiasedly, including nonlinear statistical functionals. Second, we propose a provably efficient algorithm, $\texttt{SF-LSVI}$, achieving a regret bound of $\tilde{O}(d_E H^{\frac{3}{2}}\sqrt{K})$ where $H$ is the horizon, $K$ is the number of episodes, and $d_E$ is the eluder dimension of a function class.

Salience-guided Ground Factor for Robust Localization of Delivery Robots in Complex Urban Environments

In urban environments for delivery robots, particularly in areas such as campuses and towns, many custom features defy standard road semantic categorizations. Addressing this challenge, our paper introduces a method leveraging Salient Object Detection (SOD) to extract these unique features, employing them as pivotal factors for enhanced robot loop closure and localization. Traditional geometric feature-based localization is hampered by fluctuating illumination and appearance changes. Our preference for SOD over semantic segmentation sidesteps the intricacies of classifying a myriad of non-standardized urban features. To achieve consistent ground features, the Motion Compensate IPM (MC-IPM) technique is implemented, capitalizing on motion for distortion compensation and subsequently selecting the most pertinent salient ground features through moment computations. For thorough evaluation, we validated the saliency detection and localization performances to the real urban scenarios. Project page: https://sites.google.com/view/salient-ground-feature/home.

Mesh-based Photorealistic and Real-time 3D Mapping for Robust Visual Perception of Autonomous Underwater Vehicle

This paper proposes a photorealistic real-time dense 3D mapping system that utilizes a learning-based image enhancement method and mesh-based map representation. Due to the characteristics of the underwater environment, where problems such as hazing and low contrast occur, it is hard to apply conventional simultaneous localization and mapping (SLAM) methods. Furthermore, for sensitive tasks like inspecting cracks, photorealistic mapping is very important. However, the behavior of Autonomous Underwater Vehicle (AUV) is computationally constrained. In this paper, we utilize a neural network-based image enhancement method to improve pose estimation and mapping quality and apply a sliding window-based mesh expansion method to enable lightweight, fast, and photorealistic mapping. To validate our results, we utilize real-world and indoor synthetic datasets. We performed qualitative validation with the real-world dataset and quantitative validation by modeling images from the indoor synthetic dataset as underwater scenes.

On the Convergence of Continual Learning with Adaptive Methods

One of the objectives of continual learning is to prevent catastrophic forgetting in learning multiple tasks sequentially, and the existing solutions have been driven by the conceptualization of the plasticity-stability dilemma. However, the convergence of continual learning for each sequential task is less studied so far. In this paper, we provide a convergence analysis of memory-based continual learning with stochastic gradient descent and empirical evidence that training current tasks causes the cumulative degradation of previous tasks. We propose an adaptive method for nonconvex continual learning (NCCL), which adjusts step sizes of both previous and current tasks with the gradients. The proposed method can achieve the same convergence rate as the SGD method when the catastrophic forgetting term which we define in the paper is suppressed at each iteration. Further, we demonstrate that the proposed algorithm improves the performance of continual learning over existing methods for several image classification tasks.

SPQR: Controlling Q-ensemble Independence with Spiked Random Model for Reinforcement Learning

Alleviating overestimation bias is a critical challenge for deep reinforcement learning to achieve successful performance on more complex tasks or offline datasets containing out-of-distribution data. In order to overcome overestimation bias, ensemble methods for Q-learning have been investigated to exploit the diversity of multiple Q-functions. Since network initialization has been the predominant approach to promote diversity in Q-functions, heuristically designed diversity injection methods have been studied in the literature. However, previous studies have not attempted to approach guaranteed independence over an ensemble from a theoretical perspective. By introducing a novel regularization loss for Q-ensemble independence based on random matrix theory, we propose spiked Wishart Q-ensemble independence regularization (SPQR) for reinforcement learning. Specifically, we modify the intractable hypothesis testing criterion for the Q-ensemble independence into a tractable KL divergence between the spectral distribution of the Q-ensemble and the target Wigner's semicircle distribution. We implement SPQR in several online and offline ensemble Q-learning algorithms. In the experiments, SPQR outperforms the baseline algorithms in both online and offline RL benchmarks.