arXiv:2602.23449v1 Announce Type: new Abstract: We propose an extension of the classical susceptible infectious recovered (SIR) model that incorporates the effects of spatial propagation of an epidemic through a small number of additional compartments. The model is designed to capture the dynamics of disease spread across multiple interconnected cities or populated regions, while avoiding the high dimensionality and large parameter sets typical of network based or agent-based approaches. Instead of explicitly modeling individual locations or mobility networks, we […]

Ver mais

arXiv:2601.05157v1 Announce Type: cross Abstract: In this work, we give a ${rm poly}(d,k)$ time and sample algorithm for efficiently learning the parameters of a mixture of $k$ spherical distributions in $d$ dimensions. Unlike all previous methods, our techniques apply to heavy-tailed distributions and include examples that do not even have finite covariances. Our method succeeds whenever the cluster distributions have a characteristic function with sufficiently heavy tails. Such distributions include the Laplace distribution but crucially exclude Gaussians. All […]

Ver mais

arXiv:2601.05272v1 Announce Type: new Abstract: In 1969 Strassen showed surprisingly that it is possible to multiply two 2 x 2 matrices using seven multiplications and 18 additions, instead of the naive eight multiplications and four additions. The number of additions was later reduced to 15. Karstadt and Schwartz further reduced the number of additions to 12 using a change-of-basis method. Both the number of multiplications and the number of additions have been shown to be optimal for the […]

Ver mais

arXiv:2602.14029v1 Announce Type: new Abstract: Iterative self-training (self-distillation) repeatedly refits a model on pseudo-labels generated by its own predictions. We study this procedure in overparameterized linear regression: an initial estimator is trained on noisy labels, and each subsequent iterate is trained on fresh covariates with noiseless pseudo-labels from the previous model. In the high-dimensional regime, we derive deterministic-equivalent recursions for the prediction risk and effective noise across iterations, and prove that the empirical quantities concentrate sharply around these […]

Ver mais

Cooperation in multi-agent reinforcement learning (MARL) benefits from inter-agent communication, yet most approaches assume idealized channels and existing value decomposition methods ignore who successfully shared information with whom. We propose CLOVER, a cooperative MARL framework whose centralized value mixer is conditioned on the communication graph realized under a realistic wireless channel. This graph introduces a relational inductive bias into value decomposition, constraining how individual utilities are mixed based on the realized communication structure. The mixer is a GNN […]

Ver mais

Introduction Disclaimer: This article is a cross post from Pytorch Medium Blog Post. One of the most exciting advancements, that has pushed the frontier of the Artificial Intelligence (AI) in recent years, is Deep Reinforcement Learning (DRL). DRL belongs to the family of machine learning algorithms. It assumes that intelligent machines can learn from their actions similar to the way humans learn from experience. Over the recent years we could witness some impressive real-world applications of DRL. The […]

Ver mais

We demonstrate that the fine structure constant alpha−1 ≈ 137.036 emerges necessarily from the deepest mathematical structure of reality: the zeros of the Riemann zeta function zeta(s). We present an exact formula connecting alpha−1 to the first four nontrivial zeros gamma_1, gamma_2, gamma_3, gamma_4 of zeta(1/2 + it). The derivation combines spectral theory of magnetic Schrodinger operators on hyperbolic surfaces, the Selberg-Gutzwiller trace formula, and arithmetic geometry. The resulting value matches the experimental CODATA 2018 value with precision […]

Ver mais

arXiv:2603.06626v1 Announce Type: new Abstract: Traditional Mixture-of-Experts (MoE) training typically proceeds without any structural priors, effectively requiring the model to simultaneously train expert weights while searching for an optimal routing policy within a vast combinatorial space. This entanglement often leads to sluggish convergence and training instabilities. This paper introduces Grouter, a preemptive routing method that by distilling high-quality structures from fully-trained MoE models and serving as a fixed router for target models. By decoupling structural optimization from weight […]

Ver mais

arXiv:2603.28858v1 Announce Type: new Abstract: Continual pre-training is widely used to adapt LLMs to target languages and domains, yet the mixture ratio of training data remains a sensitive hyperparameter that is expensive to tune: they must be fixed before training begins, and a suboptimal choice can waste weeks of compute. In this work, we propose OptiMer, which decouples ratio selection from training: we train one CPT model per dataset, extract each model’s distribution vector, which represents the parameter […]

Ver mais

arXiv:2601.00207v1 Announce Type: new Abstract: Rigorous crop counting is crucial for effective agricultural management and informed intervention strategies. However, in outdoor field environments, partial occlusions combined with inherent ambiguity in distinguishing clustered crops from individual viewpoints poses an immense challenge for image-based segmentation methods. To address these problems, we introduce a novel crop counting framework designed for exact enumeration via 3D instance segmentation. Our approach utilizes 2D images captured from multiple viewpoints and associates independent instance masks for […]

Ver mais