May 2026

Hi everyone — I’m working on an open-source environment for studying sequential decision-making in manufacturing systems. The current demo is a synthetic process-window benchmark: an agent/planner selects process settings, observes noisy quality outcomes, tracks uncertainty, and recommends the next experiment. The motivation is similar to sparse-data physical systems, where each real experiment is expensive and the goal is not just prediction, but deciding what to try next. Repo: https://github.com/programmablemanufacturing/programmable-manufacturing-lab I’d appreciate feedback from the RL community on: what […]

Ver mais

I’m Safiullah Korai, though many know me as Shahzaib. I’m a Software Engineer and full‑stack Flutter developer who started sharing my journey online not so long ago. Over time, by consistently documenting what I was learning and building, something interesting happened: my name began to show up. If you search Safiullah Korai on Google, you’ll see a knowledge panel, a quiet snapshot of my work and background. That small digital footprint now gives a strong first impression of credibility and […]

Ver mais

arXiv:2605.15688v1 Announce Type: new Abstract: Concept Activation Vectors (CAVs) are a fundamental tool for concept-based explainability in deep learning, yet their practical utility is limited by statistical instability. We analyze the stochastic nature of CAVs and the Testing with CAVs (TCAV) method, deriving the distributions of major CAV classes including PatternCAV, FastCAV, and ridge regression-based CAVs. We then identify a fundamental flaw in the standard TCAV score: its reliance on a discontinuous indicator function induces non-decaying variance in […]

Ver mais

arXiv:2605.15620v1 Announce Type: new Abstract: We study risk-aware offline policy learning, aiming to learn a decision rule from logged data that is optimal under general risk criteria. This problem is crucial in high-stakes domains where online interaction is infeasible and adverse outcomes must be carefully controlled. However, existing literature on offline contextual bandits either centers on expected-reward criteria or restricts risk considerations to policy evaluation instead of optimization. In this work, we propose a unified distributional framework for […]

Ver mais

arXiv:2605.15571v1 Announce Type: new Abstract: Estimating the number of distinct elements in a data stream is well understood when repeated elements are identical. In modern settings, however, observations are high-dimensional and noisy, so repeated instances of the same object are only approximately similar — for example, different images of the same individual may vary significantly at the pixel level. Classical sketches such as HyperLogLog rely on consistent hash values for identical elements and break down in this regime. […]

Ver mais

arXiv:2605.15411v1 Announce Type: new Abstract: We study contextual dynamic pricing in a semiparametric scalar-index valuation model where the latent value is $v_t=mu_ast(mathsf c_t)+xi_t$, with an unknown utility map $mu_ast$ and an unknown additive noise distribution. The key decision object is the one-dimensional oracle price map $umapsto p^ast(u)$ induced by the scalar index $u=mu_ast(mathsf c)$ and the noise tail. Under the $beta$-H”older smoothness of the tail function for $betageq 2$ and a revenue-geometry condition that gives a unique, stable, […]

Ver mais

arXiv:2605.15240v1 Announce Type: new Abstract: This paper investigates the critical role of eigenalignments between the kernel matrix and learning targets in achieving robust generalization in learning problems. We establish a direct connection between generalization performance in kernel methods and the estimation of eigenvectors and eigenvalues of matrices, offering a more intuitive understanding compared to prior work with minimal assumptions. We also show that, since the prediction task in KRR is essentially the weighted sum of eigenvectors/singular vectors, by […]

Ver mais

arXiv:2605.15206v1 Announce Type: new Abstract: Autonomous agents powered by large language models (LLMs) are increasingly used to automate complex, multi-step tasks such as coding or web-based question answering. While remote, cloud-based agents offer scalability and ease of deployment, they raise privacy concerns, depend on network connectivity, and incur recurring API costs. Deploying agents locally on user devices mitigates these issues by preserving data privacy and eliminating usage-based fees. However, agentic workflows are far more resource-intensive than typical LLM […]

Ver mais