adaptive-agents

Self-improving agents, governed. Areev is the substrate for adaptive agents — agents that get better from their own history, under human authority, in steps you can inspect, undo, and re-measure.

Latent Behavioral State Machines / Statistical Manifolds — Research

Chain-of-density study of 'Polymorphic Combinatorial Frameworks (PCF): Adaptive AI Agents' (Pearl, Murphy & Intriligator 2025, arXiv:2508.01581) - five-tier note, locator-verified. SPARK parameter spaces, topos theory, 1.25M Monte Carlo simulations.

Biologically-inspired Dual-Q reinforcement learning with asymmetric memory decay. MSc AI dissertation, University of Sussex 2025. Forgetting rewards faster than punishments improves adaptation in dynamic environments.

Adaptive reinforcement learning combat environment with curriculum learning, reward shaping, Gymnasium, FastAPI, React dashboard, and Qwen LoRA integration.

Aegis-Grid is an open-source, agentic cryptographic framework for high-assurance power system communication. It utilizes adaptive (n, k) thresholding, Shannon entropy mirroring, and Cumulative Attack Pressure (CAP) monitoring to provide a "Fail-Secure" communication backbone for critical grid infrastructure.

PUBLIC | FCTUC DEI/LEI 2020/2021 - Licenciatura em Engenharia Informatica | FIA - Fundamentos de Inteligência Artificial | Trabalhos: 1. Agentes Reativos; 2. Agentes Adaptativos