arXiv subjects

Mingxuan Zhang

Publications and source records attributed to Mingxuan Zhang.

27 records · Page 2Linked to original sources

arXiv2025

TooBadRL: Trigger Optimization to Boost Effectiveness of Backdoor Attacks on Deep Reinforcement Learning

Deep reinforcement learning (DRL) has achieved remarkable success in a wide range of sequential decision-making applications, including robotics, healthcare, smart grids, and finance. Recent studies reveal that adversaries can implant backdoors into DRL agents during the training phase. These backdoors can later be activated by specific triggers during deployment, compelling the agent to execute targeted actions and potentially leading to severe consequences, such as drone crashes or vehicle collisions. However, existing backdoor attacks utilize simplistic and heuristic trigger configurations, overlooking the critical impact of trigger design on attack effectiveness. To address this gap, we introduce TooBadRL, the first framework to systematically optimize DRL backdoor triggers across three critical aspects: injection timing, trigger dimension, and manipulation magnitude. Specifically, we first introduce a performance-aware adaptive freezing mechanism to determine the injection timing during training. Then, we formulate trigger selection as an influence attribution problem and apply Shapley value analysis to identify the most influential trigger dimension for injection. Furthermore, we propose an adversarial input synthesis method to optimize the manipulation magnitude under environmental constraints. Extensive evaluations on three DRL algorithms and nine benchmark tasks demonstrate that TooBadRL outperforms five baseline methods in terms of attack success rate while only slightly affecting normal task performance. We further evaluate potential defense strategies from detection and mitigation perspectives. We open-source our code to facilitate reproducibility and further research.

cs.CR↗

arXiv2025

MoE-SpeQ: Speculative Quantized Decoding with Proactive Expert Prefetching and Offloading for Mixture-of-Experts

The immense memory requirements of state-of-the-art Mixture-of-Experts (MoE) models present a significant challenge for inference, often exceeding the capacity of a single accelerator. While offloading experts to host memory is a common solution, it introduces a severe I/O bottleneck over the PCIe bus, as the data-dependent nature of expert selection places these synchronous transfers directly on the critical path of execution, crippling performance. This paper argues that the I/O bottleneck can be overcome by trading a small amount of cheap, on-device computation to hide the immense cost of data movement. We present MoE-SpeQ, a new inference system built on a novel co-design of speculative execution and expert offloading. MoE-SpeQ employs a small, on-device draft model to predict the sequence of required experts for future tokens. This foresight enables a runtime orchestrator to prefetch these experts from host memory, effectively overlapping the expensive I/O with useful computation and hiding the latency from the critical path. To maximize performance, an adaptive governor, guided by an Amortization Roofline Model, dynamically tunes the speculation strategy to the underlying hardware. Our evaluation on memory-constrained devices shows that for the Phi-MoE model, MoE-SpeQ achieves at most 2.34x speedup over the state-of-the-art offloading framework. Our work establishes a new, principled approach for managing data-dependent memory access in resource-limited environments, making MoE inference more accessible on commodity hardware.

cs.LG↗

arXiv2025

Energy Guided Geometric Flow Matching

A useful inductive bias for temporal data is that trajectories should stay close to the data manifold. Traditional flow matching relies on straight conditional paths, and flow matching methods which learn geodesics rely on RBF kernels or nearest neighbor graphs that suffer from the curse of dimensionality. We propose to use score matching and annealed energy distillation to learn a metric tensor that faithfully captures the underlying data geometry and informs more accurate flows. We demonstrate the efficacy of this strategy on synthetic manifolds with analytic geodesics, and interpolation of cell

cs.LG↗

arXiv2024

Magnitude Pruning of Large Pretrained Transformer Models with a Mixture Gaussian Prior

Large pretrained transformer models have revolutionized modern AI applications with their state-of-the-art performance in natural language processing (NLP). However, their substantial parameter count poses challenges for real-world deployment. To address this, researchers often reduce model size by pruning parameters based on their magnitude or sensitivity. Previous research has demonstrated the limitations of magnitude pruning, especially in the context of transfer learning for modern NLP tasks. In this paper, we introduce a new magnitude-based pruning algorithm called mixture Gaussian prior pruning (MGPP), which employs a mixture Gaussian prior for regularization. MGPP prunes non-expressive weights under the guidance of the mixture Gaussian prior, aiming to retain the model's expressive capability. Extensive evaluations across various NLP tasks, including natural language understanding, question answering, and natural language generation, demonstrate the superiority of MGPP over existing pruning methods, particularly in high sparsity settings. Additionally, we provide a theoretical justification for the consistency of the sparse transformer, shedding light on the effectiveness of the proposed pruning method.

stat.ML↗

arXiv2024

One-dimensional Photonic Crystal Structure Enhanced External-Magnetic-Field-Free Spintronic Terahertz High-Field Emitter

Intense terahertz (THz) radiation in free space offers multifaceted capabilities for accelerating electron, understanding the mesoscale architecture in (bio)materials, elementary excitation and so on. Recently popularized spintronic THz emitters (STEs) with their versatility such as ultra-broadband, cost-effectiveness, large-size and ease for-integration have become one of the most promising alternative for the next generation of intense THz sources. Nevertheless, the typical W| Co20Fe60B20 | Pt necessitates an external-magnetic-field to saturate magnetization for stable operation, limiting its scalability for achieving higher THz field with uniform distribution over larger sample areas. Here we demonstrate the methodologies of enhancing the high-field THz radiation of external-magnetic-field-free IrMn3 | Co20Fe60B20 |W heterostructure via optimizing the substrate with superior thermal conductivity and integrating a one-dimensional photonic crystal (PC) structure to maximize the radiation efficiency. Under the excitation of a Ti: sapphire femtosecond laser amplifier with central wavelength of 800 nm, pulse duration of 35 fs, and repetition rate of 1 kHz and maximum single pulse energy of 5.5 mJ, we successfully generate intense THz radiation with focal peak electric field up to 1.1 MV/cm with frequency range covering 0.1-10 THz without external-magnetic-fields. These high-field STEs will also enable other applications such as ultra-broadband high-field THz spectroscopy and polarization-based large-size strong-field THz imaging.

physics.optics↗

arXiv2023

Sparse Deep Learning for Time Series Data: Theory and Applications

Sparse deep learning has become a popular technique for improving the performance of deep neural networks in areas such as uncertainty quantification, variable selection, and large-scale network compression. However, most existing research has focused on problems where the observations are independent and identically distributed (i.i.d.), and there has been little work on the problems where the observations are dependent, such as time series data and sequential data in natural language processing. This paper aims to address this gap by studying the theory for sparse deep learning with dependent data. We show that sparse recurrent neural networks (RNNs) can be consistently estimated, and their predictions are asymptotically normally distributed under appropriate assumptions, enabling the prediction uncertainty to be correctly quantified. Our numerical results show that sparse deep learning outperforms state-of-the-art methods, such as conformal predictions, in prediction uncertainty quantification for time series data. Furthermore, our results indicate that the proposed method can consistently identify the autoregressive order for time series data and outperform existing methods in large-scale model compression. Our proposed method has important practical implications in fields such as finance, healthcare, and energy, where both accurate point estimates and prediction uncertainty quantification are of concern.

stat.ML↗

arXiv2023

Curriculum Knowledge Switching for Pancreas Segmentation

Pancreas segmentation is challenging due to the small proportion and highly changeable anatomical structure. It motivates us to propose a novel segmentation framework, namely Curriculum Knowledge Switching (CKS) framework, which decomposes detecting pancreas into three phases with different difficulty extent: straightforward, difficult, and challenging. The framework switches from straightforward to challenging phases and thereby gradually learns to detect pancreas. In addition, we adopt the momentum update parameter updating mechanism during switching, ensuring the loss converges gradually when the input dataset changes. Experimental results show that different neural network backbones with the CKS framework achieved state-of-the-art performance on the NIH dataset as measured by the DSC metric.

cs.CV↗

arXiv2022

Graph Federated Learning with Hidden Representation Sharing

Learning on Graphs (LoG) is widely used in multi-client systems when each client has insufficient local data, and multiple clients have to share their raw data to learn a model of good quality. One scenario is to recommend items to clients with limited historical data and sharing similar preferences with other clients in a social network. On the other hand, due to the increasing demands for the protection of clients' data privacy, Federated Learning (FL) has been widely adopted: FL requires models to be trained in a multi-client system and restricts sharing of raw data among clients. The underlying potential data-sharing conflict between LoG and FL is under-explored and how to benefit from both sides is a promising problem. In this work, we first formulate the Graph Federated Learning (GFL) problem that unifies LoG and FL in multi-client systems and then propose sharing hidden representation instead of the raw data of neighbors to protect data privacy as a solution. To overcome the biased gradient problem in GFL, we provide a gradient estimation method and its convergence analysis under the non-convex objective. In experiments, we evaluate our method in classification tasks on graphs. Our experiment shows a good match between our theory and the practice.

cs.LG↗

arXiv2020

Evidence for increasing frequency of extreme coastal sea levels

Projections of extreme sea levels (ESLs) are critical for managing coastal risks, but are made complicated by deep uncertainties. One key uncertainty is the choice of model structure used to estimate coastal hazards. Differences in model structural choices contribute to uncertainty in estimated coastal hazard, so it is important to characterize how model structural choice affects estimates of ESL. Here, we present a collection of 36 ESL data sets, from tide gauge stations along the United States East and Gulf Coasts. The data are processed using both annual block maxima and peaks-over-thresholds approaches for modeling distributions of extremes. We use these data sets to fit a suite of potentially nonstationary extreme value models by covarying the ESL statistics with multiple climate variables. We demonstrate how this data set enables inquiry into deep uncertainty surrounding coastal hazards. For all of the models and sites considered here, we find that accounting for changes in the frequency of coastal extreme sea levels provides a better fit than using a stationary extreme value model.

physics.data-an↗