I wanted to compare Vanilla DQN, DDQN and Dueling DDQN beyond just final reward, so I built a structured training and evaluation setup around LunarLander-v3. Instead of tracking only episode return, I monitored: • activation and gradient distributions • update-to-data ratios for optimizer diagnostics • action gap and Q-value dynamics • win rate with 95% CI intervals • generalization via human-prefix rollouts The strongest model (<9k params) achieves 98.4% win rate (±0.24%, 95% CI) across 10k seeds. The […]

Ver mais

arXiv:2602.18582v1 Announce Type: new Abstract: When training artificial intelligence (AI) to perform tasks, humans often care not only about whether a task is completed but also how it is performed. As AI agents tackle increasingly complex tasks, aligning their behavior with human-provided specifications becomes critical for responsible AI deployment. Reward design provides a direct channel for such alignment by translating human expectations into reward functions that guide reinforcement learning (RL). However, existing methods are often too limited to […]

Ver mais

What are the different ways to use and finetune pretrained large language models (LLMs)? The three most common ways to use and finetune pretrained LLMs include a feature-based approach, in-context prompting, and updating a subset of the model parameters. First, most pretrained LLMs or language transformers can be utilized without the need for further finetuning. For instance, we can employ a feature-based method to train a new downstream model, such as a linear classifier, using embeddings generated by […]

Ver mais

Physical dynamical systems can be viewed as natural information processors: their systems preserve, transform, and disperse input information. This perspective motivates learning not only from data generated by such systems, but also how to measure them in a way that extracts the most useful information for a given task. We propose a general computing framework for adaptive information extraction from dynamical systems, in which a trainable attention module learns both where to probe the system state and how […]

Ver mais

arXiv:2603.02420v1 Announce Type: new Abstract: Pluralistic alignment has emerged as a promising approach for ensuring that large language models (LLMs) faithfully represent the diversity, nuance, and conflict inherent in human values. In this work, we study a high-stakes deployment context – mulching – where automated systems transform selected individuals into nutrient-rich slurry for the dual purposes of food security and aesthetic population management. Building on recent pluralistic alignment frameworks, we introduce ValueMulch, a reproducible training, deployment, and certification […]

Ver mais

Large language model (LLM)-based multi-agent systems have demonstrated remarkable capabilities in collaborative task solving. Although the mechanisms that facilitate seamless cooperation, such as shared contexts, role assignments, and iterative message passing, present significant risks of unintentional information disclosure. We present MIN-Trust, a trust orchestration framework that enforces Minimum Necessary Information (MNI) constraints, an operationalization of the data minimization principle for inter-agent communication—while maintaining task effectiveness. Our approach introduces an MNI-Gate that automatically classifies and filters information into essential, […]

Ver mais

arXiv:2601.10869v1 Announce Type: new Abstract: This paper develops a generalized finite horizon recursive solution to the discrete time stage bound disturbance attenuation regulator (StDAR) for state feedback control. This problem addresses linear dynamical systems subject to stage bound disturbances, i.e., disturbance sequences constrained independently at each time step through stagewise squared two-norm bounds. The term generalized indicates that the results accommodate arbitrary initial states. By combining game theory and dynamic programming, this work derives a recursive solution for […]

Ver mais

Building AI features is straightforward until you need to integrate them into a production system. In production, a simple ‘call an LLM’ often becomes: wiring HTTP/JMS/Kafka/file inputs, enforcing timeouts, retries, fallbacks, stitching context obtained from internal data sources, protecting secrets, making it testable without burning API credits, and ensuring the system is observable when issues occur overnight. This is where Apache Camel excels. With Camel 4.5 and later, you can handle LLM calls as standard integration endpoints using […]

Ver mais

arXiv:2601.08954v1 Announce Type: new Abstract: This chapter examines how data analytics can be leveraged to enhance immersive teacher simulations, situating this inquiry within the broader learning sciences discourse on embodied cognition, data-informed feedback, and teacher professional learning. It explores both conceptual foundations and empirical cases to illustrate how analytics serve as mediational tools that connect immersive experiences with reflective teaching practice. The chapter unfolds in multiple sections: (1) The Innovation Journey: An Overview of Immersive Teacher Simulations outlines […]

Ver mais