arXiv subjects

Bin Du

Publications and source records attributed to Bin Du.

10 recordsLinked to original sources

arXiv2026

CANN Bench: Benchmarking Agent Generated Kernels against Real NPU and Algorithmic Limits

AI agents are now capable of writing, compiling, and iteratively optimizing low-level operator kernels on different hardware platforms. Existing benchmarks, however, focus almost exclusively on CUDA and Triton, leaving hardware ecosystems with less-exposed programming models without a common evaluation baseline. We present CANN Bench, an open benchmark for AI-generated operator code on Huawei's Ascend NPU. The current release covers 53 operators and 1060 test cases organized into four difficulty tiers -- from simple elementwise primitives to MoE dispatch and FlashAttention kernels -- spanning FP16, BF16, FP32, and INT8 precision formats. Evaluation adopts a \textbf{three-dimensional weighted composite score} that treats compilation, functional correctness, and performance as independent axes, providing a principled reward signal for kernel-generation agents. Performance is graded against an out-of-the-box PyTorch-on-Ascend baseline and an analytical per-case Hardware-Anchored Performance (HAP) limit on real NPU hardware, ensuring scores reflect genuine optimization headroom rather than measurement artifacts. The evaluation harness is designed to resist reward hacking from the ground up. CANN Bench is versioned within the official CANN repository and is designed for long-term community co-construction, providing the Ascend ecosystem with a quantitative, reproducible, and sustainably maintained yardstick for AI operator-authoring capability.

cs.AI↗

arXiv2026

Distributed Online Bandit Submodular Maximization with Bounded Sampling Violations

We study distributed online submodular maximization under partition matroid constraints, in which multiple agents select a limited number of actions from their own subsets sequentially to maximize the cumulative value of a sequence of objective functions. We develop a unified algorithmic framework that accommodates full-information and bandit feedback models. For both feedback models, we prove that the proposed algorithms achieve sublinear $(1-1/e)$-regret guarantees, which are comparable to those achieved by existing centralized counterparts. Furthermore, to tackle the sampling violation issue caused by continuous relaxation and rounding, we develop a bounded stochastic pipage rounding scheme and show that the probability of sampling violation vanishes asymptotically. As a result, the cumulative sampling violation remains sublinear in $T$, which is further shown to be not improvable under certain conditions. Numerical results validate the theoretical findings in this paper.

cs.LG↗

arXiv2026

Online Learning of Kalman Filtering: From Output to State Estimation

In this paper, we study the problem of learning Kalman filtering with unknown system model in partially observed linear dynamical systems. We propose a unified algorithmic framework based on online optimization that can be used to solve both the output estimation and state estimation scenarios. By exploring the properties of the estimation error cost functions, such as conditionally strong convexity, we show that our algorithm achieves a $\log T$-regret in the horizon length $T$ for the output estimation scenario. More importantly, we tackle the more challenging scenario of learning Kalman filtering for state estimation, which is an open problem in the literature. We first characterize a fundamental limitation of the problem, demonstrating the impossibility of any algorithm to achieve sublinear regret in $T$. By further introducing a random query scheme into our algorithm, we show that a $\sqrt{T}$-regret is achievable when rendering the algorithm limited query access to more informative measurements of the system state in practice. Our algorithm and regret readily capture the trade-off between the number of queries and the achieved regret, and shed light on online learning problems with limited observations. We validate the performance of our algorithms using numerical examples.

cs.LG↗

arXiv2024

Decentralized EM Algorithm for Gaussian Mixtures under Data Heterogeneity and Partial Labeling

We systematically study several network-based Expectation-Maximization (EM) algorithms for the Gaussian mixture model within decentralized federated learning (DFL). Our theoretical investigation shows that directly extending the classic EM algorithm to DFL leads to a biased estimator when data are heterogeneously distributed across sites. To address this, we introduce a momentum network EM (MNEM) algorithm, which integrates information from both current and historical estimators from previous DFL iterations. We further develop a semi-supervised MNEM (semi-MNEM) algorithm, which utilizes information provided by partially labeled data. Rigorous theoretical analysis demonstrates that the MNEM estimator can achieve the same asymptotic efficiency as the whole-sample estimator under appropriate regularity conditions, even with heterogeneous data. Moreover, the semi-MNEM estimator significantly improves the convergence speed of the MNEM algorithm, even if different mixture components are poorly separated. Extensive simulations are conducted, and a widely used chest X-ray dataset is analyzed to demonstrate the finite-sample performance of the proposed methods.

stat.ML↗

arXiv2024

Neural Quantile Optimization for Edge-Cloud Networking

We seek the best traffic allocation scheme for the edge-cloud computing network that satisfies constraints and minimizes the cost based on burstable billing. First, for a fixed network topology, we formulate a family of integer programming problems with random parameters describing the various traffic demands. Then, to overcome the difficulty caused by the discrete feature of the problem, we generalize the Gumbel-softmax reparameterization method to induce an unconstrained continuous optimization problem as a regularized continuation of the discrete problem. Finally, we introduce the Gumbel-softmax sampling network to solve the optimization problems via unsupervised learning. The network structure reflects the edge-cloud computing topology and is trained to minimize the expectation of the cost function for unconstrained continuous optimization problems. The trained network works as an efficient traffic allocation scheme sampler, remarkably outperforming the random strategy in feasibility and cost function value. Besides testing the quality of the output allocation scheme, we examine the generalization property of the network by increasing the time steps and the number of users. We also feed the solution to existing integer optimization solvers as initial conditions and verify the warm-starts can accelerate the short-time iteration process. The framework is general with solid performance, and the decoupled feature of the random neural networks is adequate for practical implementations.

cs.NI↗

arXiv2024

Interactive Prompt Debugging with Sequence Salience

We present Sequence Salience, a visual tool for interactive prompt debugging with input salience methods. Sequence Salience builds on widely used salience methods for text classification and single-token prediction, and extends this to a system tailored for debugging complex LLM prompts. Our system is well-suited for long texts, and expands on previous work by 1) providing controllable aggregation of token-level salience to the word, sentence, or paragraph level, making salience over long inputs tractable; and 2) supporting rapid iteration where practitioners can act on salience results, refine prompts, and run salience on the new output. We include case studies showing how Sequence Salience can help practitioners work with several complex prompting strategies, including few-shot, chain-of-thought, and constitutional principles. Sequence Salience is built on the Learning Interpretability Tool, an open-source platform for ML model visualizations, and code, notebooks, and tutorials are available at http://goo.gle/sequence-salience.

cs.CL↗

arXiv2023

Learning to Seek: Multi-Agent Online Source Seeking Against Non-Stochastic Disturbances

This paper proposes to leverage the emerging~learning techniques and devise a multi-agent online source {seeking} algorithm under unknown environment. Of particular significance in our problem setups are: i) the underlying environment is not only unknown, but dynamically changing and also perturbed by two types of non-stochastic disturbances; and ii) a group of agents is deployed and expected to cooperatively seek as many sources as possible. Correspondingly, a new technique of discounted Kalman filter is developed to tackle with the non-stochastic disturbances, and a notion of confidence bound in polytope nature is utilized~to aid the computation-efficient cooperation among~multiple agents. With standard assumptions on the unknown environment as well as the disturbances, our algorithm is shown to achieve sub-linear regrets under the two~types of non-stochastic disturbances; both results are comparable to the state-of-the-art. Numerical examples on a real-world pollution monitoring application are provided to demonstrate the effectiveness of our algorithm.

eess.SY↗

arXiv2021

Hypothesis Testing of One-Sample Mean Vector in Distributed Frameworks

Distributed frameworks are widely used to handle massive data, where sample size $n$ is very large, and data are often stored in $k$ different machines. For a random vector $X\in \mathbb{R}^p$ with expectation $μ$, testing the mean vector $H_0: μ=μ_0$ vs $H_1: μ\ne μ_0$ for a given vector $μ_0$ is a basic problem in statistics. The centralized test statistics require heavy communication costs, which can be a burden when $p$ or $k$ is large. To reduce the communication cost, distributed test statistics are proposed in this paper for this problem based on the divide and conquer technique, a commonly used approach for distributed statistical inference. Specifically, we extend two commonly used centralized test statistics to the distributed ones, that apply to low and high dimensional cases, respectively. Comparing the power of centralized test statistics and the distributed ones, it is observed that there is a fundamental tradeoff between communication costs and the powers of the tests. This is quite different from the application of the divide and conquer technique in many other problems such as estimation, where the associated distributed statistics can be as good as the centralized ones. Numerical results confirm the theoretical findings.

stat.ME↗

arXiv2021

Multi-Robot Dynamical Source Seeking in Unknown Environments

This paper presents an algorithmic framework for the distributed on-line source seeking, termed as 'DoSS', with a multi-robot system in an unknown dynamical environment. Our algorithm, building on a novel concept called dummy confidence upper bound (D-UCB), integrates both estimation of the unknown environment and task planning for the multiple robots simultaneously, and as a result, drives the team of robots to a steady state in which multiple sources of interest are located. Unlike the standard UCB algorithm in the context of multi-armed bandits, the introduction of D-UCB significantly reduces the computational complexity in solving subproblems of the multi-robot task planning. This also enables our 'DoSS' algorithm to be implementable in a distributed on-line manner. The performance of the algorithm is theoretically guaranteed by showing a sub-linear upper bound of the cumulative regret. Numerical results on a real-world methane emission seeking problem are also provided to demonstrate the effectiveness of the proposed algorithm.

cs.RO↗

arXiv2020

Jacobi-Style Iteration for Distributed Submodular Maximization

This paper presents a novel Jacobi-style iteration algorithm for solving the problem of distributed submodular maximization, in which each agent determines its own strategy from a finite set so that the global submodular objective function is jointly maximized. Building on the multi-linear extension of the global submodular function, we expect to achieve the solution from a probabilistic, rather than deterministic, perspective, and thus transfer the considered problem from a discrete domain into a continuous domain. Since it is observed that an unbiased estimation of the gradient of multi-linear extension function~can be obtained by sampling the agents' local decisions, a projected stochastic gradient algorithm is proposed to solve the problem. Our algorithm enables the distributed updates among all individual agents and is proved to asymptotically converge to a desirable equilibrium solution. Such an equilibrium solution is guaranteed to achieve at least 1/2-suboptimal bound, which is comparable to the state-of-art in the literature. Moreover, we further enhance the proposed algorithm by handling the scenario in which agents' communication delays are present. The enhanced algorithmic framework admits a more realistic distributed implementation of our approach. Finally, a movie recommendation task is conducted on a real-world movie rating data set, to validate the numerical performance of the proposed algorithms.

eess.SY↗