arXiv:2603.00935v1 Announce Type: new Abstract: Latent-space Bayesian optimization (LSBO) extends Bayesian optimization to structured domains, such as molecular design, by searching in the continuous latent space of a generative model. However, most LSBO methods assume a fixed objective, whereas real design campaigns often face temporal drift (e.g., evolving preferences or shifting targets). Bringing time-varying BO into LSBO is nontrivial: drift can affect not only the surrogate, but also the latent search space geometry induced by the representation. We […]

Ver mais

Population-based learning paradigms, including evolutionary strategies, Population-Based Training (PBT), and recent model-merging methods, combine fast within-model optimisation with slower population-level adaptation. Despite their empirical success, a general mathematical description of the resulting collective training dynamics remains incomplete. We introduce a theoretical framework for neural network training based on two-time-scale population dynamics. We model a population of neural networks as an interacting agent system in which network parameters evolve through fast noisy gradient updates of SGD/Langevin type, while hyperparameters […]

Ver mais

arXiv:2602.21585v1 Announce Type: cross Abstract: Many applications seek to optimize LLM outputs at test time by iteratively proposing, scoring, and refining candidates over a discrete output space. Existing methods use a calibrated scalar evaluator for the target objective to guide search, but for many tasks such scores are unavailable, too sparse, or unreliable. Pairwise comparisons, by contrast, are often easier to elicit, still provide useful signal on improvement directions, and can be obtained from the LLM itself without […]

Ver mais

The current development of artificial intelligence creativity tools is rapidly progressing, with Google Mixboard being another entrant. The Mixboard platform functions as an AI-driven idea board that enables creators to brainstorm ideas and create visual elements while organizing their thoughts within a single workspace. The process of developing new ideas becomes difficult for creators, marketers, and designers because they need to switch between different tools that offer inspiration, design, and execution capabilities. Google Mixboard promises to bring all […]

Ver mais

arXiv:2603.23926v1 Announce Type: cross Abstract: Online reinforcement learning in infinite-horizon Markov decision processes (MDPs) remains less theoretically and algorithmically developed than its episodic counterpart, with many algorithms suffering from high “burn-in” costs and failing to adapt to benign instance-specific complexity. In this work, we address these shortcomings for two infinite-horizon objectives: the classical average-reward regret and the $gamma$-regret. We develop a single tractable UCB-style algorithm applicable to both settings, which achieves the first optimal variance-dependent regret guarantees. Our […]

Ver mais

As companies grow, they can expand their offerings, reach new customers and markets, enhance profitability, and position themselves ahead of competitors. Whether trying to expand their company size by adding new departments or increase revenue by improving productivity per employee, companies can achieve growth through various strategies. One of the strongest company growth strategies is retaining skilled employees who drive productivity, innovation, morale, and customer satisfaction. Employee retention is a company’s ability to maintain a stable workforce by […]

Ver mais

arXiv:2604.02438v1 Announce Type: new Abstract: The deployment of reinforcement learning (RL)-based controllers on physical systems is often limited by poor generalization to real-world scenarios, known as the simulation-to-reality (sim-to-real) gap. This gap is particularly challenging in spaceflight, where real-world training data are scarce due to high cost and limited planetary exploration data. Traditional approaches, such as system identification and synthetic data generation, depend on sufficient data and often fail due to modeling assumptions or lack of physics-based constraints. […]

Ver mais

Federated learning (FL) and federated distillation (FD) are distributed learning paradigms that train UE models with enhanced privacy, each offering different trade-offs between noise robustness and learning speed. To mitigate their respective weaknesses, we propose a hybrid federated learning (HFL) framework in which each user equipment (UE) transmits either gradients or logits, and the base station (BS) selects the per-round weights of FL and FD updates. We derive convergence of HFL framework and introduce two methods to exploit […]

Ver mais

arXiv:2603.25857v1 Announce Type: new Abstract: The capabilities of large language models (LLMs) have expanded beyond natural language processing to scientific prediction tasks, including molecular property prediction. However, their effectiveness in in-context learning remains ambiguous, particularly given the potential for training data contamination in widely used benchmarks. This paper investigates whether LLMs perform genuine in-context regression on molecular properties or rely primarily on memorized values. Furthermore, we analyze the interplay between pre-trained knowledge and in-context information through a series […]

Ver mais