We establish a global finite-sample guarantee for synchronous quantile temporal-difference learning (QTD) in tabular distributional reinforcement learning. The proof separates two stability mechanisms. A global comparison argument, based on the order monotonicity of reward cumulative distribution functions and the $W_\...
Zijie Cheng, Xiang Li, Yang Peng et al.· 0 citations
A local asymptotic theory is developed for the estimated return CDF over $T^{-1/2}$-neighborhoods of finitely many thresholds, together with its bootstrap analogue that justifies bootstrap inference for smooth statistical functionals, including variance, CVaR, expected shortfall, and expectiles.
The method proposed leverages invertible residual neural networks (i-ResNets) to equip generalized linear models with both nonlinear parameter estimation and a flexible correction of their distributional assumptions while always retaining stochastic monotonicity of the modeled distribution in the (formerly linear) pred...
Tom A. Splittgerber, Niklas Koenen, Marvin N. Wright et al.· 0 citations
In experiments across open-ended generation, factual QA, and mathematical reasoning, it is shown that VCM consistently mitigates the likelihood trap, and integrates with existing decoding strategies, improving diversity and coherence and, particularly at higher decoding temperatures, reasoning accuracy.
Yuanhao Ding, Meimingwei Li, E. Arias et al.· arXiv.org· 2 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
The results show that robustness claims based on a single behavioral or representational metric can be misleading, and motivate multi-level evaluation of how perturbations alter language-model computation.
Dun-Li Chan, Emily Liu, Niyathi Allu et al.· 0 citations
Understanding how the structure of language can be learned from sentences alone is a central question in both cognitive science and machine learning. Studies of the internal representations of Large Language Models (LLMs) support their ability to parse text when predicting the next word, while representing semantic not...
Jack T. Parley, Francesco Cagnetta, Matthieu Wyart· 0 citations
SmoothFBO is proposed, the first algorithm for non-stationary FBO with both theoretical guarantees and practical scalability, and introduces a time-smoothed stochastic hypergradient estimator that reduces variance through a window parameter, enabling stable outer-loop updates with sublinear regret.
Jason Bohne, Ieva Petrulionyte, Michael Arbel et al.· arXiv.org· 2 citations
This work establishes a probabilistic version of parametric operator inference, casting the learning problem as a Bayesian linear regression, and designs a sequential adaptive sampling scheme to select new training parameter vectors that promote ROM stability and accuracy globally in the parameter domain.
Shane A. Mcquarrie, Meng-Wu Guo, A. Chaudhuri· Journal of Scientific Comput...· 1 citation
We study the problem of selecting the best heterogeneous treatment effect (HTE) estimator from a collection of candidates in settings where the treatment effect is fundamentally unobserved. We cast estimator selection as a multiple testing problem and introduce a ground-truth-free procedure based on a cross-fitted, exp...
This work proposes kernel-based methods for identifying and forecasting Hamiltonian systems directly from trajectory data, and provides a more general, problem-agnostic numerical framework that goes beyond Hamiltonian systems and can be used for data-driven learning of arbitrary dynamical systems.
Yasamin Jalalian, Mostafa Samir, B. Hamzi et al.· arXiv.org· 4 citations
We study counterfactual regression, which maps features to outcomes under hypothetical scenarios that differ from those observed in the data. This problem is central to decision-making under distribution shift, where treatment patterns may change at deployment. We develop a semiparametric framework for counterfactual r...
Activation-Keyed Momentum (AK-Momentum), which builds direction-awareness into the momentum update rule, is proposed, and it is proved that it is a valid momentum, that it applies the input-side curvature correction without matrix inversion, and that it clears stale directions faster than EMA under both a fixed and a d...
A machine-learning approach called CheMeleon provides a practical way to develop new drugs or other molecules with much less detailed, expensive experimentation.
Is your company’s AI sovereign? A practical guide for CIOs on AI vendor lock-in, EU Data Act switching, cloud sovereignty, and testing your exit plan before you need it. The post Sovereignty Is the Option to Leave appeared first on GPT-Lab.
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
