arXiv subjects

Nabuat Zaman Nahim

Publications and source records attributed to Nabuat Zaman Nahim.

6 recordsLinked to original sources

arXiv2026

Flow Matching with Missing Data

Flow matching assumes fully observed training data, which many real-world applications rarely provide. We study Missing-Data Flow Matching, which treats the missing coordinates of training samples as latent variables and averages the flow matching loss over the values they could take. We first prove that the oracle correction is exact: completing from the true pattern-conditioned posterior recovers the complete-data flow-matching objective under any missingness mechanism. Under missing completely at random (MCAR) and missing at random (MAR), this posterior equals the ordinary complete-data conditional, while under missing not at random (MNAR) it generally depends on the missingness mechanism and may require additional structure to identify. Our finite-sample analysis then answers design questions that the algorithm leaves open. Missingness transfers estimator variance rather than adding it, and for the fixed-parameter loss estimator one completion per example matches complete-data variance and is optimal under a fixed evaluation budget. A learned completion model contributes population completion bias, which we bound by its expected conditional Wasserstein distance to the true completion law. Experiments numerically validate the theoretical predictions, show that deterministic rather than frozen imputation reduces generated conditional variability, and place our method alongside strong classical and deep imputation baselines on real tabular data.

cs.LG↗

arXiv2026

Flow Matching for Offline Reinforcement Learning with Discrete Actions

Generative policies based on diffusion models and flow matching have shown strong promise for offline reinforcement learning (RL), but their applicability remains largely confined to continuous action spaces. To address a broader range of offline RL settings, we extend flow matching to a general framework that supports discrete action spaces with multiple objectives. Specifically, we replace continuous flows with continuous-time Markov chains, trained using a Q-weighted flow matching objective. We then extend our design to multi-agent settings, mitigating the exponential growth of joint action spaces via a factorized conditional path. We theoretically show that, under idealized conditions, optimizing this objective recovers the optimal policy. Extensive experiments further demonstrate that our method performs robustly across diverse settings and benchmarks, including high-dimensional control, multi-agent games, and dynamically changing preferences over multiple objectives, while outperforming traditional offline RL methods in practical multi-modal decision-making scenarios. Our discrete framework can also be applied to continuous-control problems through action quantization, providing a flexible trade-off between representational complexity and performance.

cs.LG↗

arXiv2026

Discrete Flow Matching for Offline-to-Online Reinforcement Learning

Many reinforcement learning (RL) tasks have discrete action spaces, but most generative policy methods based on diffusion and flow matching are designed for continuous control. Meanwhile, generative policies usually rely heavily on offline datasets and offline-to online RL is itself challenging, as the policy must improve from new interaction without losing useful behavior learned from static data. To address those challenges, we introduce DRIFT, an online fine-tuning method that updates an offline pretrained continuous time Markov chain (CTMC) policy with an advantage-weighted discrete flow matching loss. To preserve useful pretrained knowledge, we add a path-space penalty that regularizes the full CTMC trajectory distribution, rather than only the final action distribution. For large discrete action spaces, we introduce a candidate-set approximation that updates the actor over a small subset of actions sampled from reference-policy rollouts and uniform exploration. Our theoretical analysis shows that focusing on a small set of high probability actions can reduce computation while keeping the approximation error small, making our method practical for large action spaces. We evaluate on macro-action MinAtar games where offline-pretrained DRIFT outperforms its online-only counterpart across all five games and achieves the highest mean return on three and on oracle free Jericho, DRIFT achieves a mean normalized score of 7.3\% with a lightweight GRU encoder, outperforming the evaluated pretrained language-model baselines. Ablations demonstrate the importance of appropriately weighted path-space regularization and reference-guided candidate construction.

cs.LG↗

arXiv2026

Discrete MeanFlow: One-Step Generation via Conditional Transition Kernels

MeanFlow enables one-step generation in continuous spaces by learning an average velocity over a time interval rather than the instantaneous velocity field of flow matching. However, discrete state spaces lack smooth trajectories or spatial derivatives, so the continuous formulation does not directly apply. We introduce Discrete MeanFlow, a theoretical finite-state formulation that replaces point motion with probability-mass transport. Our key object is the conditional transition kernel of a continuous-time Markov chain (CTMC), from which we define a mean discrete rate that measures average transition-probability change over a finite interval. We prove a Discrete MeanFlow identity that relates this finite-interval rate to the instantaneous CTMC generator at the endpoint, using the Kolmogorov forward equation in place of the spatial chain rule in continuous MeanFlow. We parameterize the transition kernel directly using a boundary-by-construction design that guarantees valid probability outputs and exact boundary conditions without auxiliary losses. Because the learned kernel defines a probability distribution, generation requires only a single forward pass followed by direct sampling, eliminating the need for iterative denoising, ODE integration, or multi-step refinement. We validate the identity on exact finite-state Markov chains, demonstrating that the learned kernel recovers the analytical ground truth with high precision. We also investigate the training objective and the limits of factorization on synthetic sequence tasks across varying alphabet sizes and sequence lengths.

cs.LG↗

arXiv2026

A Survey of Flow Matching in Reinforcement Learning

Flow Matching (FM) has recently emerged as a principled and efficient generative modeling framework for reinforcement learning (RL), enabling expressive, multimodal policy parameterizations via deterministic probability transport. Compared to diffusion-based policies that rely on stochastic denoising chains, FM uses sampling based on ordinary differential equations (ODEs), with learned velocity fields, which can substantially reduce inference latency and simplify the incorporation of RL objectives. As research in flow-based RL rapidly accelerates across offline continuous control, online fine-tuning, and foundation model alignment, the literature has become highly fragmented. In this survey, we provide a comprehensive taxonomy of flow-matching approaches in reinforcement learning. We organize the literature along two axes: the target distribution being modeled (e.g., action policies, value critics, transition dynamics) and the mechanism of RL signal integration (e.g., energy-weighted regression, flow-based policy gradients, and group relative policy optimization). Furthermore, we survey emerging frontiers such as discrete and non-Euclidean action spaces, provide a systematic comparative analysis against Gaussian and diffusion baselines, and outline critical open problems. Ultimately, this survey serves as a foundational roadmap for the next generation of generative reinforcement learning and alignment.

cs.LG↗

arXiv2024

Enhancing Bidirectional Sign Language Communication: Integrating YOLOv8 and NLP for Real-Time Gesture Recognition & Translation

The primary concern of this research is to take American Sign Language (ASL) data through real time camera footage and be able to convert the data and information into text. Adding to that, we are also putting focus on creating a framework that can also convert text into sign language in real time which can help us break the language barrier for the people who are in need. In this work, for recognising American Sign Language (ASL), we have used the You Only Look Once(YOLO) model and Convolutional Neural Network (CNN) model. YOLO model is run in real time and automatically extracts discriminative spatial-temporal characteristics from the raw video stream without the need for any prior knowledge, eliminating design flaws. The CNN model here is also run in real time for sign language detection. We have introduced a novel method for converting text based input to sign language by making a framework that will take a sentence as input, identify keywords from that sentence and then show a video where sign language is performed with respect to the sentence given as input in real time. To the best of our knowledge, this is a rare study to demonstrate bidirectional sign language communication in real time in the American Sign Language (ASL).

cs.CV↗