We study the stochastic linear bandits with heavy-tailed noise. Two principled strategies for handling heavy-tailed noise, truncation and median-of-means, have been introduced to heavy-tailed bandits. Nonetheless, these methods rely on specific noise assumptions or bandit structures, limiting their applicability to general settings. The recent work [Huang et al.2024] develops a soft truncation method via the adaptive Huber regression to address these limitations. However, their method suffers undesired computational cost: it requires storing all historical data and performing a full pass over these data at each round. In this paper, we propose a \emph{one-pass} algorithm based on the online mirror descent framework. Our method updates using only current data at each round, reducing the per-round computational cost from $\widetilde{\mathcal{O}}(t \log T)$ to $\widetilde{\mathcal{O}}(1)$ with respect to current round $t$ and the time horizon $T$, and achieves a near-optimal and variance-aware regret of order $\widetilde{\mathcal{O}}\big(d T^{\frac{1-\epsilon}{2(1+\epsilon)}} \sqrt{\sum_{t=1}^T \nu_t^2} + d T^{\frac{1-\epsilon}{2(1+\epsilon)}}\big)$ where $d$ is the dimension and $\nu_t^{1+\epsilon}$ is the $(1+\epsilon)$-th central moment of reward at round $t$.

Reinforcement Learning from Human Feedback is a key technique for training large language models using human feedback [Ouyang et al., 2022, Bai et al., 2022]. The RLHF process involves collecting data, each consisting of a prompt, a pair of responses, and a human preference label indicating the preferred response. Then, a reward model is trained to predict human preferences, and the LLM is fine-tuned based on the reward model by RL algorithms, such as PPO [Schulman et al., 2017]. Given the notable success of RLHF, recent efforts have been devoted to developing a deeper theoretical understanding of this approach. Zhu et al. [2023] investigated the standard offline setting, where the learner is provided with a fixed dataset and aims to learn a policy that maximizes the expected reward. In this setting, since the learner has no control over the data collection process, the quality of the dataset becomes crucial to the performance of the learned policy and the resulting policy often performs poorly when faced with out-of-distribution data [Burns et al., 2024]. In practice, the Claude [Bai et al., 2022] and LLaMA-2 [Touvron et al., 2023] projects have demonstrated that iterative RLHF can significantly enhance model performance.

Recent advances in neural models have shown considerable promise in solving Traveling Salesman Problems (TSPs) without relying on much hand-crafted engineering. However, while non-autoregressive (NAR) approaches benefit from faster inference through parallelism, they typically deliver solutions of inferior quality compared to autoregressive ones. To enhance the solution quality while maintaining fast inference, we propose DEITSP, a diffusion model with efficient iterations tailored for TSP that operates in a NAR manner. Firstly, we introduce a one-step diffusion model that integrates the controlled discrete noise addition process with self-consistency enhancement, enabling optimal solution prediction through simultaneous denoising of multiple solutions. Secondly, we design a dual-modality graph transformer to bolster the extraction and fusion of features from node and edge modalities, while further accelerating the inference with fewer layers. Thirdly, we develop an efficient iterative strategy that alternates between adding and removing noise to improve exploration compared to previous diffusion methods. Additionally, we devise a scheduling framework to progressively refine the solution space by adjusting noise levels, facilitating a smooth search for optimal solutions. Extensive experiments on real-world and large-scale TSP instances demonstrate that DEITSP performs favorably against existing neural approaches in terms of solution quality, inference latency, and generalization ability. Our code is available at $\href{https://github.com/DEITSP/DEITSP}{https://github.com/DEITSP/DEITSP}$.

We study episodic linear mixture MDPs with the unknown transition and adversarial rewards under full-information feedback, employing dynamic regret as the performance measure. We start with in-depth analyses of the strengths and limitations of the two most popular methods: occupancy-measure-based and policy-based methods. We observe that while the occupancy-measure-based method is effective in addressing non-stationary environments, it encounters difficulties with the unknown transition. In contrast, the policy-based method can deal with the unknown transition effectively but faces challenges in handling non-stationary environments. Building on this, we propose a novel algorithm that combines the benefits of both methods. Specifically, it employs (i) an occupancy-measure-based global optimization with a two-layer structure to handle non-stationary environments; and (ii) a policy-based variance-aware value-targeted regression to tackle the unknown transition. We bridge these two parts by a novel conversion. Our algorithm enjoys an $\widetilde{\mathcal{O}}(d \sqrt{H^3 K} + \sqrt{HK(H + \bar{P}_K)})$ dynamic regret, where $d$ is the feature dimension, $H$ is the episode length, $K$ is the number of episodes, $\bar{P}_K$ is the non-stationarity measure. We show it is minimax optimal up to logarithmic factors by establishing a matching lower bound. To the best of our knowledge, this is the first work that achieves near-optimal dynamic regret for adversarial linear mixture MDPs with the unknown transition without prior knowledge of the non-stationarity measure.

The emergence of Large Language Models (LLMs) specifically designed for code-related tasks has marked a significant advancement in code generation. The code LLMs (Roziere et al., 2023; Zheng et al., 2023; Guo et al., 2024a; Hui et al., 2024) pre-trained on extensive datasets comprising billions of code-related tokens further revolutionize the automation of software development tasks, providing contextually relevant code suggestions and facilitating the translation from natural language to code. The generation capability of code LLMs opens up diverse applications in software development, promising to enhance productivity and streamline coding processes. As the field continues to evolve, it presents exciting opportunities for future developments and innovations in automated programming and code assistance. The code completion task is crucial in modern software development, enhancing coding efficiency and accuracy by predicting and suggesting code segments based on context. Recent advancements in code LLMs (Bavarian et al., 2022b) have introduced sophisticated completion techniques, such as prefix-suffix-middle (PSM) and suffix-prefix-middle (SPM) paradigms, which can complete middle code segments given the surrounding context. However, the current benchmark (Ding et al., 2024; Liu et al., 2023a) mainly focuses on several programming languages. For example, the Cross-CodeEval (Ding et al., 2024) includes four languages (i.e., Python, Java, TypeScript, C#). Besides, existing benchmarks can only provide the average score among all samples, which can not provide a language-specific evaluation for different programming languages based on their intrinsic structure.

Out-of-distribution (OOD) detection is crucial for deploying robust machine learning models, especially in areas where security is critical. However, traditional OOD detection methods often fail to capture complex data distributions from large scale date. In this paper, we present a novel approach for OOD detection that leverages the generative ability of diffusion models and the powerful feature extraction capabilities of CLIP. By using these features as conditional inputs to a diffusion model, we can reconstruct the images after encoding them with CLIP. The difference between the original and reconstructed images is used as a signal for OOD identification. The practicality and scalability of our method is increased by the fact that it does not require class-specific labeled ID data, as is the case with many other methods. Extensive experiments on several benchmark datasets demonstrates the robustness and effectiveness of our method, which have significantly improved the detection accuracy.

Real-time video analytics systems typically place models with fewer weights on edge devices to reduce latency. The distribution of video content features may change over time for various reasons (i.e. light and weather change) , leading to accuracy degradation of existing models, to solve this problem, recent work proposes a framework that uses a remote server to continually train and adapt the lightweight model at edge with the help of complex model. However, existing analytics approaches leave two challenges untouched: firstly, retraining task is compute-intensive, resulting in large model update delays; secondly, new model may not fit well enough with the data distribution of the current video stream. To address these challenges, in this paper, we present EdgeSync, EdgeSync filters the samples by considering both timeliness and inference results to make training samples more relevant to the current video content as well as reduce the update delay, to improve the quality of training, EdgeSync also designs a training management module that can efficiently adjusts the model training time and training order on the runtime. By evaluating real datasets with complex scenes, our method improves about 3.4% compared to existing methods and about 10% compared to traditional means.

Code completion models have made significant progress in recent years. Recently, repository-level code completion has drawn more attention in modern software development, and several baseline methods and benchmarks have been proposed. However, existing repository-level code completion methods often fall short of fully using the extensive context of a project repository, such as the intricacies of relevant files and class hierarchies. Besides, the existing benchmarks usually focus on limited code completion scenarios, which cannot reflect the repository-level code completion abilities well of existing methods. To address these limitations, we propose the R2C2-Coder to enhance and benchmark the real-world repository-level code completion abilities of code Large Language Models, where the R2C2-Coder includes a code prompt construction method R2C2-Enhance and a well-designed benchmark R2C2-Bench. Specifically, first, in R2C2-Enhance, we first construct the candidate retrieval pool and then assemble the completion prompt by retrieving from the retrieval pool for each completion cursor position. Second, based on R2C2 -Enhance, we can construct a more challenging and diverse R2C2-Bench with training, validation and test splits, where a context perturbation strategy is proposed to simulate the real-world repository-level code completion well. Extensive results on multiple benchmarks demonstrate the effectiveness of our R2C2-Coder.

To address the uncertainty in function types, recent progress in online convex optimization (OCO) has spurred the development of universal algorithms that simultaneously attain minimax rates for multiple types of convex functions. However, for a $T$-round online problem, state-of-the-art methods typically conduct $O(\log T)$ projections onto the domain in each round, a process potentially time-consuming with complicated feasible sets. In this paper, inspired by the black-box reduction of Cutkosky and Orabona (2018), we employ a surrogate loss defined over simpler domains to develop universal OCO algorithms that only require $1$ projection. Embracing the framework of prediction with expert advice, we maintain a set of experts for each type of functions and aggregate their predictions via a meta-algorithm. The crux of our approach lies in a uniquely designed expert-loss for strongly convex functions, stemming from an innovative decomposition of the regret into the meta-regret and the expert-regret. Our analysis sheds new light on the surrogate loss, facilitating a rigorous examination of the discrepancy between the regret of the original loss and that of the surrogate loss, and carefully controlling meta-regret under the strong convexity condition. In this way, with only $1$ projection per round, we establish optimal regret bounds for general convex, exponentially concave, and strongly convex functions simultaneously. Furthermore, we enhance the expert-loss to exploit the smoothness property, and demonstrate that our algorithm can attain small-loss regret for multiple types of convex and smooth functions.

We study a new class of MDPs that employs multinomial logit (MNL) function approximation to ensure valid probability distributions over the state space. Despite its benefits, introducing non-linear function approximation raises significant challenges in both computational and statistical efficiency. The best-known method of Hwang and Oh [2023] has achieved an $\widetilde{\mathcal{O}}(\kappa^{-1}dH^2\sqrt{K})$ regret, where $\kappa$ is a problem-dependent quantity, $d$ is the feature space dimension, $H$ is the episode length, and $K$ is the number of episodes. While this result attains the same rate in $K$ as the linear cases, the method requires storing all historical data and suffers from an $\mathcal{O}(K)$ computation cost per episode. Moreover, the quantity $\kappa$ can be exponentially small, leading to a significant gap for the regret compared to the linear cases. In this work, we first address the computational concerns by proposing an online algorithm that achieves the same regret with only $\mathcal{O}(1)$ computation cost. Then, we design two algorithms that leverage local information to enhance statistical efficiency. They not only maintain an $\mathcal{O}(1)$ computation cost per episode but achieve improved regrets of $\widetilde{\mathcal{O}}(\kappa^{-1/2}dH^2\sqrt{K})$ and $\widetilde{\mathcal{O}}(dH^2\sqrt{K} + \kappa^{-1}d^2H^2)$ respectively. Finally, we establish a lower bound, justifying the optimality of our results in $d$ and $K$. To the best of our knowledge, this is the first work that achieves almost the same computational and statistical efficiency as linear function approximation while employing non-linear function approximation for reinforcement learning.