Searcharxiv⌕ Search

arXiv subjects

Dongmei Chen

Publications and source records attributed to Dongmei Chen.

32 records · Page 2Linked to original sources
arXiv2025

Natural Geometry of Robust Data Attribution: From Convex Models to Deep Networks

Data attribution methods identify which training examples are responsible for a model's predictions, but their sensitivity to distributional perturbations undermines practical reliability. We present a unified framework for certified robust attribution that extends from convex models to deep networks. For convex settings, we derive Wasserstein-Robust Influence Functions (W-RIF) with provable coverage guarantees. For deep networks, we demonstrate that Euclidean certification is rendered vacuous by spectral amplification -- a mechanism where the inherent ill-conditioning of deep representations inflates Lipschitz bounds by over $10{,}000\times$. This explains why standard TRAK scores, while accurate point estimates, are geometrically fragile: naive Euclidean robustness analysis yields 0\% certification. Our key contribution is the Natural Wasserstein metric, which measures perturbations in the geometry induced by the model's own feature covariance. This eliminates spectral amplification, reducing worst-case sensitivity by $76\times$ and stabilizing attribution estimates. On CIFAR-10 with ResNet-18, Natural W-TRAK certifies 68.7\% of ranking pairs compared to 0\% for Euclidean baselines -- to our knowledge, the first non-vacuous certified bounds for neural network attribution. Furthermore, we prove that the Self-Influence term arising from our analysis equals the Lipschitz constant governing attribution stability, providing theoretical grounding for leverage-based anomaly detection. Empirically, Self-Influence achieves 0.970 AUROC for label noise detection, identifying 94.1\% of corrupted labels by examining just the top 20\% of training data.

cs.LG↗
arXiv2025

Cosine-Similarity Methods for Efficient Training and Sampling in High-Dimensional Latent Spaces

Latent generative models are increasingly shifting from traditional VAEs toward representation autoencoders and semantically aligned latent spaces, which lift images into higher-dimensional feature domains where semantic factors become more separable. Yet these spaces also contain geometric regularities that existing methods do not fully exploit--particularly in the directional relationships between features. We introduce a cosine-similarity-based mechanism that improves both training and sampling by selecting couplings that produce cleaner, less entangled velocity fields. This simple alignment reduces gradient noise, accelerates convergence, and improves sample fidelity. Building on this idea, we develop cosine-similarity-based fine-tuning and time-scheduling strategies that reduce the FID of an 800-epoch RAE from 11.99 to 8.60. Furthermore, by formulating an optimal-transport coupling using a cosine cost, a single-epoch fine-tuning step at the 20-epoch checkpoint reaches 3.30 FID-matching the performance of the 80-epoch baseline.

math.OC↗
arXiv2025

Algorithm-Relative Trajectory Valuation in Policy Gradient Control

We study how trajectory value depends on the learning algorithm in policy-gradient control. Using Trajectory Shapley in an uncertain LQR, we find a negative correlation between Persistence of Excitation (PE) and marginal value under vanilla REINFORCE ($r\approx-0.38$). We prove a variance-mediated mechanism: (i) for fixed energy, higher PE yields lower gradient variance; (ii) near saddles, higher variance increases escape probability, raising marginal contribution. When stabilized (state whitening or Fisher preconditioning), this variance channel is neutralized and information content dominates, flipping the correlation positive ($r\approx+0.29$). Hence, trajectory value is algorithm-relative. Experiments validate the mechanism and show decision-aligned scores (Leave-One-Out) complement Shapley for pruning, while Shapley identifies toxic subsets.

cs.LG↗
arXiv2025

From Natural Language to Certified H-infinity Controllers: Integrating LLM Agents with LMI-Based Synthesis

We present \textsc{S2C} (Specification-to-Certified-Controller), a multi-agent framework that maps natural-language requirements to certified $\mathcal{H}_\infty$ state-feedback controllers via LMI synthesis. \textsc{S2C} coordinates five roles -- \textit{SpecInt} (spec extraction), \textit{Solv} (bounded-real lemma (BRL) LMI), \textit{Tester} (Monte Carlo and frequency-domain checks), \textit{Adapt} (spec refinement), and \textit{CodeGen} (deployable code). The loop is stabilized by a severity- and iteration-aware $γ$-floor guardrail and a decay-rate region constraint enforcing $\Reλ(A{+}BK)<-α$ with $α=3.9/T_s$ derived from settling-time targets. For state feedback, verification reports disturbance rejection $\big\|C\,(sI-(A{+}BK))^{-1}E\big\|_\infty$ alongside time-domain statistics; discrete benchmarks are converted to continuous time via a Tustin (bilinear) transform when needed. On 14 COMPleib problems, \textsc{S2C} attains \textbf{100\%} synthesis success and \textbf{100\%} convergence within six iterations, with strong decay-rate satisfaction and near-target certified $\mathcal{H}_\infty$ levels; it improves robustness metrics relative to single-shot BRL and BRL+$α$ baselines. An ablation over LLM backbones (GPT-5, GPT-5 mini, DeepSeek-V3, Qwen-2.5-72B, Llama-4 Maverick) shows the pipeline is robust across models, while stronger models yield the highest effectiveness. These results indicate that LLM agents can integrate certificate-bearing control synthesis from high-level intent, enabling rapid end-to-end prototyping without sacrificing formal guarantees.

eess.SY↗
arXiv2025

Trajectory-Optimized Density Control with Flow Matching

Optimal transport (OT) and Schr{ö}dinger bridge (SB) problems have emerged as powerful frameworks for transferring probability distributions with minimal cost. However, existing approaches typically focus on endpoint matching while neglecting critical path-dependent properties -- particularly collision avoidance in multiagent systems -- which limits their practical applicability in robotics, economics, and other domains where inter-agent interactions are essential. Moreover, traditional density control methods often rely on independence assumptions that fail to capture swarm dynamics. We propose a novel framework that addresses these limitations by employing flow matching as the core modeling tool, where the flow model co-evolves with the control policy. Unlike prior methods that treat transport trajectories as mere interpolations between source and target distributions, our approach explicitly optimizes over the entire transport path, enabling the incorporation of trajectory-dependent costs and collision avoidance constraints. Our framework bridges optimal transport theory with mean field control, providing a principled approach to multiagent coordination problems where both endpoint alignment and path properties are critical. Experimental results demonstrate that our method successfully generates collision-free transport plans while maintaining computational efficiency comparable to standard flow matching approaches.

math.OC↗
arXiv2025

Model Predictive Path Integral Control for Roll-to-Roll Manufacturing

Roll-to-roll (R2R) manufacturing is a continuous processing technology essential for scalable production of thin-film materials and printed electronics, but precise control remains challenging due to subsystem interactions, nonlinearities, and process disturbances. This paper proposes a Model Predictive Path Integral (MPPI) control formulation for R2R systems, leveraging a GPU-based Monte-Carlo sampling approach to efficiently approximate optimal controls online. Crucially, MPPI easily handles non-differentiable cost functions, enabling the incorporation of complex performance criteria relevant to advanced manufacturing processes. A case study is presented that demonstrates that MPPI significantly improves tension regulation performance compared to conventional model predictive control (MPC), highlighting its suitability for real-time control in advanced manufacturing.

eess.SY↗
arXiv2025

$H_\infty$ Performance Analysis for Almost Periodic Piecewise Linear Systems with Application to Roll-to-Roll Manufacturing Control

An almost periodic piecewise linear system (APPLS) is a type of piecewise linear system where the system cyclically switches between different modes, each with an uncertain but bounded dwell-time. Process regulation, especially disturbance rejection, is critical to the performance of these advanced systems. However, a method to guarantee disturbance rejection has not been developed. The objective of this study is to develop an $H_\infty$ performance analysis method for APPLSs, building on which an algorithm to synthesize practical $H_\infty$ controllers is proposed. As an application, the developed methods are demonstrated with an advanced manufacturing system -- roll-to-roll (R2R) dry transfer of two-dimensional materials and printed flexible electronics. Experimental results show that the proposed method enables a less conservative and much better performing $H_\infty$ controller compared with a baseline $H_\infty$ controller that does not account for the uncertain system switching structure.

eess.SY↗
arXiv2025

Existence and Uniqueness of Solution for Linear Complementarity Problem in Contact Mechanics

Although a unique solution is guaranteed in the Linear complementarity problem (LCP) when the matrix $\mathbf{M}$ is positive definite, practical applications often involve cases where $\mathbf{M}$ is only positive semi-definite, leading to multiple possible solutions. However, empirical observations suggest that uniqueness can still emerge under certain structural conditions on the matrix $\mathbf{M}$ and vector $\mathbf{q}$. Motivated by an unresolved problem in nonlinear modeling for beam contact in directional drilling, this paper systematically investigates conditions under which a unique solution exists for LCPs with certain positive semi-definite matrices $\mathbf{M}$. We provide a rigorous proof demonstrating the existence and uniqueness of the solution for this specific case and extend our findings to establish a generalized framework applicable to broader classes of LCPs. This framework enhances the understanding of LCP uniqueness conditions and provides theoretical guarantees for solving real-world problems where positive semi-definite matrices $\mathbf{M}$ arise.

math.OC↗
arXiv2025

MDR-DeePC: Model-Inspired Distributionally Robust Data-Enabled Predictive Control

This paper presents a Model-Inspired Distributionally Robust Data-enabled Predictive Control (MDR-DeePC) framework for systems with partially known and uncertain dynamics. The proposed method integrates model-based equality constraints for known dynamics with a Hankel matrix-based representation of unknown dynamics. A distributionally robust optimization problem is formulated to account for parametric uncertainty and stochastic disturbances. Simulation results on a triple-mass-spring-damper system demonstrate improved disturbance rejection, reduced output oscillations, and lower control cost compared to standard DeePC. The results validate the robustness and effectiveness of MDR-DeePC, with potential for real-time implementation pending further benchmarking.

eess.SY↗
arXiv2025

Constrained Optimal Planning to Minimize Battery Degradation of Autonomous Mobile Robots

This paper proposes an optimization framework that addresses both cycling degradation and calendar aging of batteries for autonomous mobile robot (AMR) to minimize battery degradation while ensuring task completion. A rectangle method of piecewise linear approximation is employed to linearize the bilinear optimization problem. We conduct a case study to validate the efficiency of the proposed framework in achieving an optimal path planning for AMRs while reducing battery aging.

cs.RO↗
arXiv2025

Robust Optimal Task Planning to Maximize Battery Life

This paper proposes a control-oriented optimization platform for autonomous mobile robots (AMRs), focusing on extending battery life while ensuring task completion. The requirement of fast AMR task planning while maintaining minimum battery state of charge, thus maximizing the battery life, renders a bilinear optimization problem. McCormick envelop technique is proposed to linearize the bilinear term. A novel planning algorithm with relaxed constraints is also developed to handle parameter uncertainties robustly with high efficiency ensured. Simulation results are provided to demonstrate the utility of the proposed methods in reducing battery degradation while satisfying task completion requirements.

cs.RO↗
arXiv2023

SLO/GO Degradation-Loss Sensitivity in Climate-Human System Coupling

The potential of extreme environmental change driven by a destabilized climate system is an alarming prospect for humanity. But the intricate, subtle ways Earth's climate couples to social and economic systems raise the question of when more incremental climate change signals the need for alarm. Questions about incremental sensitivity are particularly crucial for human systems that are organized by optimization. Optimization is most valuable in resolving complex interactions among multiple factors, however, those interactions can obscure coupling to underlying drivers such as environmental degradation. Here, using Multi-Objective Land Allocation as an example, we show that model features that are common across non-convex optimization problems drive hypersensitivities in climate-induced degradation--loss response. We show that catastrophic losses in human systems can occur well before catastrophic climate collapse. We find punctuated insensitive/hypersensitive degradation--loss response, which we trace to the contrasting effects of environmental degradation on subleading, local versus global optima (SLO/GO). We argue that the SLO/GO response we identify in land-allocation problems traces to features that are common across non-convex optimization problems more broadly. Given the broad range of human systems that rely on non-convex optimization, our results therefore suggest that substantial social and economic risks could be lurking in a broad range in human systems that are coupled to the environment, even in the absence of catastrophic changes to the environment itself.

cs.CE↗
arXiv2023

Flashpoints Signal Hidden Inherent Instabilities in Land-Use Planning

Land-use decision-making processes have a long history of producing globally pervasive systemic equity and sustainability concerns. Quantitative, optimization-based planning approaches, e.g. Multi-Objective Land Allocation (MOLA), seemingly open the possibility to improve objectivity and transparency by explicitly evaluating planning priorities by the type, amount, and location of land uses. Here, we show that optimization-based planning approaches with generic planning criteria generate a series of unstable "flashpoints" whereby tiny changes in planning priorities produce large-scale changes in the amount of land use by type. We give quantitative arguments that the flashpoints we uncover in MOLA models are examples of a more general family of instabilities that occur whenever planning accounts for factors that coordinate use on- and between-sites, regardless of whether these planning factors are formulated explicitly or implicitly. We show that instabilities lead to regions of ambiguity in land-use type that we term "gray areas". By directly mapping gray areas between flashpoints, we show that quantitative methods retain utility by reducing combinatorially large spaces of possible land-use patterns to a small, characteristic set that can engage stakeholders to arrive at more efficient and just outcomes.

cs.AI↗