arXiv subjects

Arnob Ghosh

Publications and source records attributed to Arnob Ghosh.

At least 19 recordsLinked to original sources

arXiv2026

Robust Peak-cost Constrained Reinforcement Learning

We study robust peak-cost constrained reinforcement learning (RP-CRL), where the objective is to maximize expected reward while controlling the maximum cost encountered along a trajectory. This setting is motivated by safety-critical applications in which a single large violation can be catastrophic and therefore cannot be adequately captured by the standard CMDP framework based on expected cumulative cost. Existing reachability-constrained RL methods adopt Lagrangian-based approaches, yet the underlying duality properties of peak-cost constrained MDPs remain unclear. We show that, unlike standard CMDPs, peak-cost constrained MDPs may not admit zero duality gap. We further consider a robust formulation to address simulator-to-real-world mismatch in the transition dynamics. To solve this problem, we develop a surrogate optimization framework and a robust value estimation method based on integral probability metrics. We prove that, with appropriate hyperparameter choices, the surrogate solution attains the same robust reward value as the original problem while violating the constraint by at most epsilon. Experiments show that the proposed method effectively enforces safety under dynamics perturbations while retaining strong reward performance.

cs.LG↗

arXiv2026

Revisiting Subgradient Dominance in Robust MDPs: Counterexamples, Hardness, and Sufficient Conditions

Projected subgradient descent (PSD) has gained popularity for solving robust Markov decision processes (RMDPs) because it applies to a broader class of uncertainty sets than traditional dynamic programming. Existing work claims that RMDPs with a general compact uncertainty set satisfy the subgradient dominance property, under which exact PSD converges to an $\varepsilon$-optimal policy in a polynomial number of updates (e.g., Wang et al., 2023). We show that these claims are incorrect. Even when the uncertainty set has cardinality two, the RMDP objective is not subgradient-dominant and can admit suboptimal strict local minima. Moreover, we prove that finding an $\varepsilon$-optimal policy can be NP-hard even in settings where subgradients are efficiently computable: (i) finite transition uncertainty sets and (ii) $sa$-rectangular finite transition uncertainty sets with finite cost uncertainty sets. Finally, we identify two conditions under which RMDPs do satisfy subgradient dominance: when, for each policy, either the worst-case transition kernel or the worst-case action-value function is unique.

math.OC↗

arXiv2026

Optimistic Policy Learning under Pessimistic Adversaries with Regret and Violation Guarantees

Real-world decision-making systems operate in environments where state transitions depend not only on the agent's actions, but also on \textbf{exogenous factors outside its control}--competing agents, environmental disturbances, or strategic adversaries--formally, $s_{h+1} = f(s_h, a_h, \bar{a}_h)+\omega_h$ where $\bar{a}_h$ is the adversary/external action, $a_h$ is the agent's action, and $\omega_h$ is an additive noise. Ignoring such factors can yield policies that are optimal in isolation but \textbf{fail catastrophically in deployment}, particularly when safety constraints must be satisfied. Standard Constrained MDP formulations assume the agent is the sole driver of state evolution, an assumption that breaks down in safety-critical settings. Existing robust RL approaches address this via distributional robustness over transition kernels, but do not explicitly model the \textbf{strategic interaction} between agent and exogenous factor, and rely on strong assumptions about divergence from a known nominal model. We model the exogenous factor as an \textbf{adversarial policy} $\bar{\pi}$ that co-determines state transitions, and ask how an agent can remain both optimal and safe against such an adversary. \emph{To the best of our knowledge, this is the first work to study safety-constrained RL under explicit adversarial dynamics}. We propose \textbf{Robust Hallucinated Constrained Upper-Confidence RL} (\texttt{RHC-UCRL}), a model-based algorithm that maintains optimism over both agent and adversary policies, explicitly separating epistemic from aleatoric uncertainty. \texttt{RHC-UCRL} achieves sub-linear regret and constraint violation guarantees.

cs.LG↗

arXiv2026

Beyond Freshness and Semantics: A Coupon-Collector Framework for Effective Status Updates

For status update systems operating over unreliable energy-constrained wireless channels, we address Weaver's long-standing Level-C question: do my packets actually improve the plant's behavior? Each fresh sample carries a stochastic expiration time -- governed by the plant's instability dynamics -- after which the information becomes useless for control. Casting the problem as a coupon-collector variant with expiring coupons, we (i) formulate a two-dimensional average-reward MDP, (ii) prove that the optimal schedule is doubly thresholded in the receiver's freshness timer and the sender's stored lifetime, (iii) derive a closed-form policy for deterministic lifetimes, and (iv) design a Structure-Aware Q-learning algorithm (SAQ) that learns the optimal policy without knowing the channel success probability or lifetime distribution. Simulations validate our theoretical predictions: SAQ matches optimal Value Iteration performance while converging significantly faster than baseline Q-learning, and expiration-aware scheduling achieves up to 50% higher reward than age-based baselines by adapting transmissions to state-dependent urgency -- thereby delivering Level-C effectiveness under tight resource constraints.

eess.SY↗

arXiv2026

Escaping Offline Pessimism: Vector-Field Reward Shaping for Safe Frontier Exploration

While offline reinforcement learning provides reliable policies for real-world deployment, its inherent pessimism severely restricts an agent's ability to explore and collect novel data online. Drawing inspiration from safe reinforcement learning, exploring near the boundary of regions well covered by the offline dataset and reliably modeled by the simulator allows an agent to take manageable risks--venturing into informative but moderate-uncertainty states while remaining close enough to familiar regions for safe recovery. However, naively rewarding this boundary-seeking behavior can lead to a degenerate parking behavior, where the agent simply stops once it reaches the frontier. To solve this, we propose a novel vector-field reward shaping paradigm designed to induce continuous, safe boundary exploration for non-adaptive deployed policies. Operating on an uncertainty oracle trained from offline data, our reward combines two complementary components: a gradient-alignment term that attracts the agent toward a target uncertainty level, and a rotational-flow term that promotes motion along the local tangent plane of the uncertainty manifold. Through theoretical analysis, we show that this reward structure naturally induces sustained exploratory behavior along the boundary while preventing degenerate solutions. Empirically, by integrating our proposed reward shaping with Soft Actor-Critic on a 2D continuous navigation task, we validate that agents successfully traverse uncertainty boundaries while balancing safe, informative data collection with primary task completion.

cs.LG↗

arXiv2026

Finding Super-spreaders in SIS Epidemics

In network epidemic models, controlling the spread of a disease often requires targeted interventions such as vaccinating high-risk individuals based on network structure. However, typical approaches assume complete knowledge of the underlying contact network, which is often unavailable. While network structure can be learned from observed epidemic dynamics, existing methods require long observation windows that may delay critical interventions. In this work, we show that full network reconstruction may not be necessary: control-relevant features, such as high-degree vertices (super-spreaders), can be learned far more efficiently than the complete structure. Specifically, we develop an algorithm to identify such vertices from the dynamics of a Susceptible-Infected-Susceptible (SIS) process. We prove that in an $n$-vertex graph, vertices of degree at least $n^\alpha$ can be identified over an observation window of size $\Omega (1/\alpha)$, for any $\alpha \in (0,1)$. In contrast, existing methods for exact network reconstruction requires an observation window that grows linearly with $n$. Simulations demonstrate that our approach accurately identifies super-spreaders and enables effective epidemic control.

math.ST↗

arXiv2025

Provably Efficient Sample Complexity for Robust CMDP

We study the problem of learning policies that maximize cumulative reward while satisfying safety constraints, even when the real environment differs from a simulator or nominal model. We focus on robust constrained Markov decision processes (RCMDPs), where the agent must maximize reward while ensuring cumulative utility exceeds a threshold under the worst-case dynamics within an uncertainty set. While recent works have established finite-time iteration complexity guarantees for RCMDPs using policy optimization, their sample complexity guarantees remain largely unexplored. In this paper, we first show that Markovian policies may fail to be optimal even under rectangular uncertainty sets unlike the {\em unconstrained} robust MDP. To address this, we introduce an augmented state space that incorporates the remaining utility budget into the state representation. Building on this formulation, we propose a novel Robust constrained Value iteration (RCVI) algorithm with a sample complexity of $\mathcal{\tilde{O}}(|S||A|H^5/\epsilon^2)$ achieving at most $\epsilon$ violation using a generative model where $|S|$ and $|A|$ denote the sizes of the state and action spaces, respectively, and $H$ is the episode length. To the best of our knowledge, this is the {\em first sample complexity guarantee} for RCMDP. Empirical results further validate the effectiveness of our approach.

cs.LG↗

arXiv2025

Certifiable Safe RLHF: Semantic Grounding and Fixed Penalty Constraint Optimization for Safer LLM Alignment

Ensuring safety is a foundational requirement for large language models (LLMs). Achieving an appropriate balance between enhancing the utility of model outputs and mitigating their potential for harm is a complex and persistent challenge. Contemporary approaches frequently formalize this problem within the framework of Constrained Markov Decision Processes (CMDPs) and employ established CMDP optimization techniques. However, these methods exhibit two notable limitations. First, their reliance on reward and cost functions renders performance highly sensitive to the underlying scoring mechanism, which must capture semantic meaning rather than being triggered by superficial keywords. Second, CMDP-based training entails tuning dual-variable, a process that is both computationally expensive and does not provide any provable safety guarantee for a fixed dual variable that can be exploitable through adversarial jailbreaks. To overcome these limitations, we introduce Certifiable Safe-RLHF (CS-RLHF) that introduces a cost model trained on a large-scale corpus to assign semantically grounded safety scores. In contrast to the lagrangian-based approach, CS-RLHF adopts a rectified penalty-based formulation. This design draws on the theory of exact penalty functions in constrained optimization, wherein constraint satisfaction is enforced directly through a suitably chosen penalty term. With an appropriately scaled penalty, feasibility of the safety constraints can be guaranteed at the optimizer, eliminating the need for dual-variable updates. Empirical evaluation demonstrates that CS-RLHF outperforms state-of-the-art LLM model responses rendering at-least 5 times efficient against nominal and jail-breaking prompts

cs.LG↗

arXiv2025

Tail-Risk-Safe Monte Carlo Tree Search under PAC-Level Guarantees

Making decisions with respect to just the expected returns in Monte Carlo Tree Search (MCTS) cannot account for the potential range of high-risk, adverse outcomes associated with a decision. To this end, safety-aware MCTS often consider some constrained variants -- by introducing some form of mean risk measures or hard cost thresholds. These approaches fail to provide rigorous tail-safety guarantees with respect to extreme or high-risk outcomes (denoted as tail-risk), potentially resulting in serious consequence in high-stake scenarios. This paper addresses the problem by developing two novel solutions. We first propose CVaR-MCTS, which embeds a coherent tail risk measure, Conditional Value-at-Risk (CVaR), into MCTS. Our CVaR-MCTS with parameter $\alpha$ achieves explicit tail-risk control over the expected loss in the "worst $(1-\alpha)\%$ scenarios." Second, we further address the estimation bias of tail-risk due to limited samples. We propose Wasserstein-MCTS (or W-MCTS) by introducing a first-order Wasserstein ambiguity set $\mathcal{P}_{\varepsilon_{s}}(s,a)$ with radius $\varepsilon_{s}$ to characterize the uncertainty in tail-risk estimates. We prove PAC tail-safety guarantees for both CVaR-MCTS and W-MCTS and establish their regret. Evaluations on diverse simulated environments demonstrate that our proposed methods outperform existing baselines, effectively achieving robust tail-risk guarantees with improved rewards and stability.

cs.LG↗

arXiv2025

Efficient Policy Optimization in Robust Constrained MDPs with Iteration Complexity Guarantees

Constrained decision-making is essential for designing safe policies in real-world control systems, yet simulated environments often fail to capture real-world adversities. We consider the problem of learning a policy that will maximize the cumulative reward while satisfying a constraint, even when there is a mismatch between the real model and an accessible simulator/nominal model. In particular, we consider the robust constrained Markov decision problem (RCMDP) where an agent needs to maximize the reward and satisfy the constraint against the worst possible stochastic model under the uncertainty set centered around an unknown nominal model. Primal-dual methods, effective for standard constrained MDP (CMDP), are not applicable here because of the lack of the strong duality property. Further, one cannot apply the standard robust value-iteration based approach on the composite value function either as the worst case models may be different for the reward value function and the constraint value function. We propose a novel technique that effectively minimizes the constraint value function--to satisfy the constraints; on the other hand, when all the constraints are satisfied, it can simply maximize the robust reward value function. We prove that such an algorithm finds a policy with at most $\epsilon$ sub-optimality and feasible policy after $O(\epsilon^{-2})$ iterations. In contrast to the state-of-the-art method, we do not need to employ a binary search, thus, we reduce the computation time by at least 4x for smaller value of discount factor ($\gamma$) and by at least 6x for larger value of $\gamma$.

cs.LG↗

arXiv2025

LIMCA: LLM for Automating Analog In-Memory Computing Architecture Design Exploration

Resistive crossbars enabling analog In-Memory Computing (IMC) have emerged as a promising architecture for Deep Neural Network (DNN) acceleration, offering high memory bandwidth and in-situ computation. However, the manual, knowledge-intensive design process and the lack of high-quality circuit netlists have significantly constrained design space exploration and optimization to behavioral system-level tools. In this work, we introduce LIMCA, a novel fine-tune-free Large Language Model (LLM)-driven framework for automating the design and evaluation of IMC crossbar architectures. Unlike traditional approaches, LIMCA employs a No-Human-In-Loop (NHIL) automated pipeline to generate and validate circuit netlists for SPICE simulations, eliminating manual intervention. LIMCA systematically explores the IMC design space by leveraging a structured dataset and LLM-based performance evaluation. Our experimental results on MNIST classification demonstrate that LIMCA successfully generates crossbar designs achieving $\geq$96% accuracy while maintaining a power consumption $\leq$3W, making this the first work in LLM-assisted IMC design space exploration. Compared to existing frameworks, LIMCA provides an automated, scalable, and hardware-aware solution, reducing design exploration time while ensuring user-constrained performance trade-offs.

cs.AR↗

arXiv2025

TPU-Gen: LLM-Driven Custom Tensor Processing Unit Generator

The increasing complexity and scale of Deep Neural Networks (DNNs) necessitate specialized tensor accelerators, such as Tensor Processing Units (TPUs), to meet various computational and energy efficiency requirements. Nevertheless, designing optimal TPU remains challenging due to the high domain expertise level, considerable manual design time, and lack of high-quality, domain-specific datasets. This paper introduces TPU-Gen, the first Large Language Model (LLM) based framework designed to automate the exact and approximate TPU generation process, focusing on systolic array architectures. TPU-Gen is supported with a meticulously curated, comprehensive, and open-source dataset that covers a wide range of spatial array designs and approximate multiply-and-accumulate units, enabling design reuse, adaptation, and customization for different DNN workloads. The proposed framework leverages Retrieval-Augmented Generation (RAG) as an effective solution for a data-scare hardware domain in building LLMs, addressing the most intriguing issue, hallucinations. TPU-Gen transforms high-level architectural specifications into optimized low-level implementations through an effective hardware generation pipeline. Our extensive experimental evaluations demonstrate superior performance, power, and area efficiency, with an average reduction in area and power of 92\% and 96\% from the manual optimization reference values. These results set new standards for driving advancements in next-generation design automation tools powered by LLMs.

cs.AR↗

arXiv2025

Provably Efficient RL for Linear MDPs under Instantaneous Safety Constraints in Non-Convex Feature Spaces

In Reinforcement Learning (RL), tasks with instantaneous hard constraints present significant challenges, particularly when the decision space is non-convex or non-star-convex. This issue is especially relevant in domains like autonomous vehicles and robotics, where constraints such as collision avoidance often take a non-convex form. In this paper, we establish a regret bound of $\tilde{\mathcal{O}}\bigl(\bigl(1 + \tfrac{1}{\tau}\bigr) \sqrt{\log(\tfrac{1}{\tau}) d^3 H^4 K} \bigr)$, applicable to both star-convex and non-star-convex cases, where $d$ is the feature dimension, $H$ the episode length, $K$ the number of episodes, and $\tau$ the safety threshold. Moreover, the violation of safety constraints is zero with high probability throughout the learning process. A key technical challenge in these settings is bounding the covering number of the value-function class, which is essential for achieving value-aware uniform concentration in model-free function approximation. For the star-convex setting, we develop a novel technique called Objective Constraint-Decomposition (OCD) to properly bound the covering number. This result also resolves an error in a previous work on constrained RL. In non-star-convex scenarios, where the covering number can become infinitely large, we propose a two-phase algorithm, Non-Convex Safe Least Squares Value Iteration (NCS-LSVI), which first reduces uncertainty about the safe set by playing a known safe policy. After that, it carefully balances exploration and exploitation to achieve the regret bound. Finally, numerical simulations on an autonomous driving scenario demonstrate the effectiveness of NCS-LSVI.

cs.LG↗

arXiv2025

Provably Efficient RL under Episode-Wise Safety in Constrained MDPs with Linear Function Approximation

We study the reinforcement learning (RL) problem in a constrained Markov decision process (CMDP), where an agent explores the environment to maximize the expected cumulative reward while satisfying a single constraint on the expected total utility value in every episode. While this problem is well understood in the tabular setting, theoretical results for function approximation remain scarce. This paper closes the gap by proposing an RL algorithm for linear CMDPs that achieves $\tilde{\mathcal{O}}(\sqrt{K})$ regret with an episode-wise zero-violation guarantee. Furthermore, our method is computationally efficient, scaling polynomially with problem-dependent parameters while remaining independent of the state space size. Our results significantly improve upon recent linear CMDP algorithms, which either violate the constraint or incur exponential computational costs.

cs.LG↗

arXiv2025

Performing Load Balancing under Constraints

Join-the-shortest queue (JSQ) and its variants have often been used in solving load balancing problems. The aim of such policies is to minimize the average system occupation, e.g., the customer's system time. In this work we extend the traditional load balancing setting to include constraints that may be imposed, e.g., due to the communication network. We cast the problem into the framework of constrained MDPs, enabling the consideration of both action-dependent constraints, such as, e.g, bandwidth limitation, and state-dependent constraints, such as, e.g., minimum queue utilization. Unlike the state-of-the-art approaches, our load-balancing policies, in particular JSED-$k$ and JSSQ, are both provably safe and yet strive to minimize the system occupancy. Their performance is tested with extensive numerical results under various system settings.

cs.NI↗

arXiv2024

SPICEPilot: Navigating SPICE Code Generation and Simulation with AI Guidance

Large Language Models (LLMs) have shown great potential in automating code generation; however, their ability to generate accurate circuit-level SPICE code remains limited due to a lack of hardware-specific knowledge. In this paper, we analyze and identify the typical limitations of existing LLMs in SPICE code generation. To address these limitations, we present SPICEPilot a novel Python-based dataset generated using PySpice, along with its accompanying framework. This marks a significant step forward in automating SPICE code generation across various circuit configurations. Our framework automates the creation of SPICE simulation scripts, introduces standardized benchmarking metrics to evaluate LLM's ability for circuit generation, and outlines a roadmap for integrating LLMs into the hardware design process. SPICEPilot is open-sourced under the permissive MIT license at https://github.com/ACADLab/SPICEPilot.git.

cs.AR↗

arXiv2024

SA-DS: A Dataset for Large Language Model-Driven AI Accelerator Design Generation

In the ever-evolving landscape of Deep Neural Networks (DNN) hardware acceleration, unlocking the true potential of systolic array accelerators has long been hindered by the daunting challenges of expertise and time investment. Large Language Models (LLMs) offer a promising solution for automating code generation which is key to unlocking unprecedented efficiency and performance in various domains, including hardware descriptive code. The generative power of LLMs can enable the effective utilization of preexisting designs and dedicated hardware generators. However, the successful application of LLMs to hardware accelerator design is contingent upon the availability of specialized datasets tailored for this purpose. To bridge this gap, we introduce the Systolic Array-based Accelerator Data Set (SA-DS). SA-DS comprises a diverse collection of spatial array designs following the standardized Berkeley's Gemmini accelerator generator template, enabling design reuse, adaptation, and customization. SA-DS is intended to spark LLM-centered research on DNN hardware accelerator architecture. We envision that SA-DS provides a framework that will shape the course of DNN hardware acceleration research for generations to come. SA-DS is open-sourced under the permissive MIT license at https://github.com/ACADLab/SA-DS.git}{https://github.com/ACADLab/SA-DS.

cs.AR↗

arXiv2024

Adversarially Trained Weighted Actor-Critic for Safe Offline Reinforcement Learning

We propose WSAC (Weighted Safe Actor-Critic), a novel algorithm for Safe Offline Reinforcement Learning (RL) under functional approximation, which can robustly optimize policies to improve upon an arbitrary reference policy with limited data coverage. WSAC is designed as a two-player Stackelberg game to optimize a refined objective function. The actor optimizes the policy against two adversarially trained value critics with small importance-weighted Bellman errors, which focus on scenarios where the actor's performance is inferior to the reference policy. In theory, we demonstrate that when the actor employs a no-regret optimization oracle, WSAC achieves a number of guarantees: (i) For the first time in the safe offline RL setting, we establish that WSAC can produce a policy that outperforms any reference policy while maintaining the same level of safety, which is critical to designing a safe algorithm for offline RL. (ii) WSAC achieves the optimal statistical convergence rate of $1/\sqrt{N}$ to the reference policy, where $N$ is the size of the offline dataset. (iii) We theoretically show that WSAC guarantees a safe policy improvement across a broad range of hyperparameters that control the degree of pessimism, indicating its practical robustness. Additionally, we offer a practical version of WSAC and compare it with existing state-of-the-art safe offline RL algorithms in several continuous control environments. WSAC outperforms all baselines across a range of tasks, supporting the theoretical results.

cs.LG↗