arXiv subjects

P. R. Kumar

Publications and source records attributed to P. R. Kumar.

At least 19 recordsLinked to original sources

arXiv2026

ReGuide: From Test-Time Guidance to Self-Improving Diffusion Policies

Behavior-cloned diffusion policies are expressive but remain vulnerable to covariate shift: small deviations from demonstrated states can compound into task failure. Existing methods address this either by expanding the training distribution through expert corrections or synthetic augmentation, or by steering a frozen policy at test time with guidance from a learned model. The former can be expensive or assumption-dependent, while the latter discards the corrected trajectories after execution. We introduce ReGuide, a self-improving framework that treats guided rollouts as reusable on-policy recovery data. ReGuide first uses Phase-Conditioned Guidance (PCG) to generate corrective rollouts: it constructs phase-specific latent targets, applies guidance only in the drifted-but-recoverable regime, and guides through the estimated clean action to match the dynamics model's training distribution. Successful guided rollouts are then absorbed back into the policy through ReGuide-FT, which fine-tunes the current checkpoint, or ReGuide-FS, which retrains from scratch on the augmented dataset; the two can also be composed and iterated. On Robomimic Can, Square, Transport, and Tool Hang, ReGuide improves base-policy success by $1.3$--$7.7\times$, outperforms LPB in the test-time-only setting, and matched-data ablations show that the gains come from guided recovery data rather than additional rollouts alone.

cs.LG↗

arXiv2026

Dreaming Smoothly and Sample Efficiently with Gradient Penalized Latent Dynamics

Model-based reinforcement learning improves sample efficiency by learning a world model. However, existing latent world models such as DreamerV3 do not explicitly enforce local smoothness in their learned transition dynamics, leaving a useful inductive bias for transition dynamics learning unexploited. We propose GPLD, a gradient-penalized latent dynamics regularizer for DreamerV3 that applies a row-wise Jacobian penalty to the posterior latent distribution to encourage locally smooth transition learning. We show that this penalty can be interpreted as the continuous-latent analog of finite-difference smoothing of transition laws in discrete embedded-state MDPs, and estimate it efficiently using Hutchinson-style stochastic probes. Empirically, across DeepMind Control proprioceptive tasks, GPLD improves aggregate sample efficiency, with particularly strong gains on higher-complexity locomotion environments. On more challenging quadruped tasks, GPLD reaches high-return behavior earlier and exhibits more consistent late-stage learning over longer horizons. Explicit local smoothness regularization is a simple and effective way to improve latent world models for smooth continuous control environments. Code for GPLD is available at github.com/romils9/gpld-mbrl .

cs.LG↗

arXiv2026

Optimistic World Models: Efficient Exploration in Model-Based Deep Reinforcement Learning

Efficient exploration remains a central challenge in reinforcement learning (RL), particularly in sparse-reward environments. We introduce Optimistic World Models (OWMs), a principled and scalable framework for optimistic exploration that brings classical reward-biased maximum likelihood estimation (RBMLE) from adaptive control into deep RL. In contrast to upper confidence bound (UCB)-style exploration methods, OWMs incorporate optimism directly into model learning by augmentation with an optimistic dynamics loss that biases imagined transitions toward higher-reward outcomes. This fully gradient-based loss requires neither uncertainty estimates nor constrained optimization. Our approach is plug-and-play with existing world model frameworks, preserving scalability while requiring only minimal modifications to standard training procedures. We instantiate OWMs within two state-of-the-art world model architectures, leading to Optimistic DreamerV3 and Optimistic STORM, which demonstrate significant improvements in sample efficiency and cumulative return compared to their baseline counterparts.

cs.LG↗

arXiv2025

A Dynamic Watermarking Technique for Matching Communication Addresses with Cars in a Visual Field

We consider a problem faced by an intelligent roadside unit (RSU) monitoring a roadway by a video camera. Suppose the RSU notices that a particular car in its visual field needs to execute a specific evasive maneuver to avoid danger. It would like to send a packet addressed to that particular car with this suggestion. The problem is that while all the cars are communicating with the RSU, the RSU does not know which car in the video is associated with what IP address. So, it does not know which IP address to send the packet to. Indeed, the problem of matching addresses with cars in the visual field is a fundamental open problem. We provide an active solution employing dynamic watermarking that was originally developed for the security of cyber-physical systems. This technique calls for a car to superpose a small random excitation onto its actuation commands for steering angle or throttle/brake positions. The car sends this random waveform to the RSU in a packet containing its IP address. By signal processing of the video stream of a car at the RSU it can verify whether it matches with the waveform in the packet and thereby associates that the IP address of the packet with that car in the visual field. The RSU thereby determines which IP address is associated with which car in its visual field. We present two demonstrations of performance. We demonstrate experimental results on a laboratory transportation automated vehicles, a vision system, and a network, as well as on the field with two passenger sedans in practice. The results demonstrate that employing the dynamic watermarking method enables an RSU to distinguish the communication of a target vehicle from that of other IP addresses of nearby vehicles.

eess.SP↗

arXiv2025

Vector preference-based contextual bandits under distributional shifts

We consider contextual bandit learning under distribution shift when reward vectors are ordered according to a given preference cone. We propose an adaptive-discretization and optimistic elimination based policy that self-tunes to the underlying distribution shift. To measure the performance of this policy, we introduce the notion of preference-based regret which measures the performance of a policy in terms of distance between Pareto fronts. We study the performance of this policy by establishing upper bounds on its regret under various assumptions on the nature of distribution shift. Our regret bounds generalize known results for the existing case of no distribution shift and vectorial reward settings, and scale gracefully with problem parameters in presence of distribution shifts.

cs.LG↗

arXiv2024

The Transmission Value of Energy Storage and Fundamental Limitations

This study addresses the transmission value of energy storage in electric grids. The inherent connection between storage and transmission infrastructure is captured from a "cumulative energy" perspective, which enables the reformulating of the conventional optimization problem by employing line power flow as the decision variable. The study also establishes the theoretical limitations of both storage and transmission lines that can be replaced by each other, providing explicit closed-form expressions for the minimum capacity needed. As a key departure from conventional practice in which transmission lines are designed according to the peak power delivery needs, with sufficient storage capacity, the transmission line capacity can be designed based on the average power delivery needs. The models of this paper only rely on a few basic assumptions, paving the way for understanding future storage as a transmission asset market design. Numerical experiments based on 2-bus, modified RTS 24-bus, RTS-GMLC, and Texas synthetic power systems illustrate the results.

eess.SY↗

arXiv2024

Linear Convergence of Independent Natural Policy Gradient in Games with Entropy Regularization

This work focuses on the entropy-regularized independent natural policy gradient (NPG) algorithm in multi-agent reinforcement learning. In this work, agents are assumed to have access to an oracle with exact policy evaluation and seek to maximize their respective independent rewards. Each individual's reward is assumed to depend on the actions of all the agents in the multi-agent system, leading to a game between agents. We assume all agents make decisions under a policy with bounded rationality, which is enforced by the introduction of entropy regularization. In practice, a smaller regularization implies the agents are more rational and behave closer to Nash policies. On the other hand, agents with larger regularization acts more randomly, which ensures more exploration. We show that, under sufficient entropy regularization, the dynamics of this system converge at a linear rate to the quantal response equilibrium (QRE). Although regularization assumptions prevent the QRE from approximating a Nash equilibrium, our findings apply to a wide range of games, including cooperative, potential, and two-player matrix games. We also provide extensive empirical results on multiple games (including Markov games) as a verification of our theoretical analysis.

cs.LG↗

arXiv2024

Provable Policy Gradient Methods for Average-Reward Markov Potential Games

We study Markov potential games under the infinite horizon average reward criterion. Most previous studies have been for discounted rewards. We prove that both algorithms based on independent policy gradient and independent natural policy gradient converge globally to a Nash equilibrium for the average reward criterion. To set the stage for gradient-based methods, we first establish that the average reward is a smooth function of policies and provide sensitivity bounds for the differential value functions, under certain conditions on ergodicity and the second largest eigenvalue of the underlying Markov decision process (MDP). We prove that three algorithms, policy gradient, proximal-Q, and natural policy gradient (NPG), converge to an $\epsilon$-Nash equilibrium with time complexity $O(\frac{1}{\epsilon^2})$, given a gradient/differential Q function oracle. When policy gradients have to be estimated, we propose an algorithm with $\tilde{O}(\frac{1}{\min_{s,a}\pi(a|s)\delta})$ sample complexity to achieve $\delta$ approximation error w.r.t~the $\ell_2$ norm. Equipped with the estimator, we derive the first sample complexity analysis for a policy gradient ascent algorithm, featuring a sample complexity of $\tilde{O}(1/\epsilon^5)$. Simulation studies are presented.

cs.LG↗

arXiv2023

Value-Biased Maximum Likelihood Estimation for Model-based Reinforcement Learning in Discounted Linear MDPs

We consider the infinite-horizon linear Markov Decision Processes (MDPs), where the transition probabilities of the dynamic model can be linearly parameterized with the help of a predefined low-dimensional feature mapping. While the existing regression-based approaches have been theoretically shown to achieve nearly-optimal regret, they are computationally rather inefficient due to the need for a large number of optimization runs in each time step, especially when the state and action spaces are large. To address this issue, we propose to solve linear MDPs through the lens of Value-Biased Maximum Likelihood Estimation (VBMLE), which is a classic model-based exploration principle in the adaptive control literature for resolving the well-known closed-loop identification problem of Maximum Likelihood Estimation. We formally show that (i) VBMLE enjoys $\widetilde{O}(d\sqrt{T})$ regret, where $T$ is the time horizon and $d$ is the dimension of the model parameter, and (ii) VBMLE is computationally more efficient as it only requires solving one optimization problem in each time step. In our regret analysis, we offer a generic convergence result of MLE in linear MDPs through a novel supermartingale construct and uncover an interesting connection between linear MDPs and online learning, which could be of independent interest. Finally, the simulation results show that VBMLE significantly outperforms the benchmark method in terms of both empirical regret and computation time.

cs.LG↗

arXiv2023

Provably Fast Convergence of Independent Natural Policy Gradient for Markov Potential Games

This work studies an independent natural policy gradient (NPG) algorithm for the multi-agent reinforcement learning problem in Markov potential games. It is shown that, under mild technical assumptions and the introduction of the \textit{suboptimality gap}, the independent NPG method with an oracle providing exact policy evaluation asymptotically reaches an $\epsilon$-Nash Equilibrium (NE) within $\mathcal{O}(1/\epsilon)$ iterations. This improves upon the previous best result of $\mathcal{O}(1/\epsilon^2)$ iterations and is of the same order, $\mathcal{O}(1/\epsilon)$, that is achievable for the single-agent case. Empirical results for a synthetic potential game and a congestion game are presented to verify the theoretical bounds.

cs.LG↗

arXiv2023

Natural Actor-Critic for Robust Reinforcement Learning with Function Approximation

We study robust reinforcement learning (RL) with the goal of determining a well-performing policy that is robust against model mismatch between the training simulator and the testing environment. Previous policy-based robust RL algorithms mainly focus on the tabular setting under uncertainty sets that facilitate robust policy evaluation, but are no longer tractable when the number of states scales up. To this end, we propose two novel uncertainty set formulations, one based on double sampling and the other on an integral probability metric. Both make large-scale robust RL tractable even when one only has access to a simulator. We propose a robust natural actor-critic (RNAC) approach that incorporates the new uncertainty sets and employs function approximation. We provide finite-time convergence guarantees for the proposed RNAC algorithm to the optimal robust policy within the function approximation error. Finally, we demonstrate the robust performance of the policy learned by our proposed RNAC approach in multiple MuJoCo environments and a real-world TurtleBot navigation task.

cs.LG↗

arXiv2023

Finite Time Regret Bounds for Minimum Variance Control of Autoregressive Systems with Exogenous Inputs

Minimum variance controllers have been employed in a wide-range of industrial applications. A key challenge experienced by many adaptive controllers is their poor empirical performance in the initial stages of learning. In this paper, we address the problem of initializing them so that they provide acceptable transients, and also provide an accompanying finite-time regret analysis, for adaptive minimum variance control of an auto-regressive system with exogenous inputs (ARX). Following [3], we consider a modified version of the Certainty Equivalence (CE) adaptive controller, which we call PIECE, that utilizes probing inputs for exploration. We show that it has a $C \log T$ bound on the regret after $T$ time-steps for bounded noise, and $C\log^2 T$ in the case of sub-Gaussian noise. The simulation results demonstrate the advantage of PIECE over the algorithm proposed in [3] as well as the standard Certainty Equivalence controller especially in the initial learning phase. To the best of our knowledge, this is the first work that provides finite-time regret bounds for an adaptive minimum variance controller.

eess.SY↗

arXiv2023

Bounded (O(1)) Regret Recommendation Learning via Synthetic Controls Oracle

In online exploration systems where users with fixed preferences repeatedly arrive, it has recently been shown that O(1), i.e., bounded regret, can be achieved when the system is modeled as a linear contextual bandit. This result may be of interest for recommender systems, where the popularity of their items is often short-lived, as the exploration itself may be completed quickly before potential long-run non-stationarities come into play. However, in practice, exact knowledge of the linear model is difficult to justify. Furthermore, potential existence of unobservable covariates, uneven user arrival rates, interpretation of the necessary rank condition, and users opting out of private data tracking all need to be addressed for practical recommender system applications. In this work, we conduct a theoretical study to address all these issues while still achieving bounded regret. Aside from proof techniques, the key differentiating assumption we make here is the presence of effective Synthetic Control Methods (SCM), which are shown to be a practical relaxation of the exact linear model knowledge assumption. We verify our theoretical bounded regret result using a minimal simulation experiment.

cs.LG↗

arXiv2023

TERRA: Beam Management for Outdoor mm-Wave Networks

mm-Wave communication systems use narrow directional beams due to the spectrum's characteristic nature: high path and penetration losses. The mobile and the base station primarily employ beams in line of sight (LoS) direction and when needed in non-line of sight direction. Beam management protocol adapts the base station and mobile side beam direction during user mobility and to sustain the link during blockages. To avoid outage in transient pedestrian blockage of the LoS path, the mobile uses reflected or NLoS path available in indoor environments. Reflected paths can sustain time synchronization and maintain connectivity during temporary blockages. In outdoor environments, such reflections may not be available and prior work relied on dense base station deployment or co-ordinated multi-point access to address outage problem. Instead of dense and hence cost-intensive network deployments, we found experimentally that the mobile can capitalize on ground reflection. We developed TERRA protocol to effectively handle mobile side beam direction during transient blockage events. TERRA avoids outage during pedestrian blockages 84.5 $\%$ of the time in outdoor environments on concrete and gravel surfaces. TERRA also enables the mobile to perform a soft handover to a reserve neighbor base station in the event of a permanent blockage, without requiring any side information, unlike the existing works. Evaluations show that TERRA maintains received signal strength close to the optimal solution while keeping track of the neighbor base station.

eess.SY↗

arXiv2022

Energy System Digitization in the Era of AI: A Three-Layered Approach towards Carbon Neutrality

The transition towards carbon-neutral electricity is one of the biggest game changers in addressing climate change since it addresses the dual challenges of removing carbon emissions from the two largest sectors of emitters: electricity and transportation. The transition to a carbon-neutral electric grid poses significant challenges to conventional paradigms of modern grid planning and operation. Much of the challenge arises from the scale of the decision making and the uncertainty associated with the energy supply and demand. Artificial Intelligence (AI) could potentially have a transformative impact on accelerating the speed and scale of carbon-neutral transition, as many decision making processes in the power grid can be cast as classic, though challenging, machine learning tasks. We point out that to amplify AI's impact on carbon-neutral transition of the electric energy systems, the AI algorithms originally developed for other applications should be tailored in three layers of technology, markets, and policy.

cs.AI↗

arXiv2022

Terra: Blockage Resilience in Outdoor mmWave Networks

We address the problem of pedestrian blockage in outdoor mm-Wave networks, which can disrupt Line of Sight (LoS) communication and result in an outage. This necessitates either reacquisition as a new user that can take up to 1.28 sec in 5G New Radio, disrupting high-performance applications, or handover to a different base station (BS), which however requires very costly dense deployments to ensure multiple base stations are always visible to a mobile outdoors. We have found that there typically exists a strong ground reflection from concrete and gravel surfaces, within 4-6 dB of received signal strength (RSS) of LoS paths. The mobile can switch to such a Non-Line Sight (NLoS) beam to sustain the link for use as a control channel during a blockage event. This allows the mobile to maintain time-synchronization with the base station, allowing it to revert to the LoS path when the temporary blockage disappears. We present a protocol, Terra, to quickly discover, cache, and employ ground reflections. It can be used in most outdoor built environments since pedestrian blockages typically last only a few hundred milliseconds.

eess.SY↗

arXiv2022

Anchor-Changing Regularized Natural Policy Gradient for Multi-Objective Reinforcement Learning

We study policy optimization for Markov decision processes (MDPs) with multiple reward value functions, which are to be jointly optimized according to given criteria such as proportional fairness (smooth concave scalarization), hard constraints (constrained MDP), and max-min trade-off. We propose an Anchor-changing Regularized Natural Policy Gradient (ARNPG) framework, which can systematically incorporate ideas from well-performing first-order methods into the design of policy optimization algorithms for multi-objective MDP problems. Theoretically, the designed algorithms based on the ARNPG framework achieve $\tilde{O}(1/T)$ global convergence with exact gradients. Empirically, the ARNPG-guided algorithms also demonstrate superior performance compared to some existing policy gradient-based approaches in both exact gradients and sample-based scenarios.

cs.LG↗

arXiv2022

On an Information and Control Architecture for Future Electric Energy Systems

This paper presents considerations towards an information and control architecture for future electric energy systems driven by massive changes resulting from the societal goals of decarbonization and electrification. This paper describes the new requirements and challenges of an extended information and control architecture that need to be addressed for continued reliable delivery of electricity. It identifies several new actionable information and control loops, along with their spatial and temporal scales of operation, which can together meet the needs of future grids and enable deep decarbonization of the electricity sector. The present architecture of electric power grids designed in a different era is thereby extensible to allow the incorporation of increased renewables and other emerging electric loads.

eess.SY↗