Machine Learning
273 articles in this category (Page 8 of 12)
Self-Supervised Temporal Pattern Mining for Wildfire Evacuation Logistics Networks Under Real-Time Policy Constraints
This article details a novel approach to wildfire evacuation logistics, leveraging self-supervised learning to improve adaptability to changing conditions, achieving a more robust system than traditional supervised methods.
Self-Supervised Temporal Pattern Mining for circular manufacturing supply chains with embodied agent feedback loops
A novel system combining self-supervised learning and embodied agents achieves a 42% improvement in predicting component return volumes in circular manufacturing.
A Coding Implementation on Building Self-Organizing Zettelkasten Knowledge Graphs and Sleep-Consolidation Mechanisms
This tutorial demonstrates building a “Zettelkasten” memory system using Gemini, achieving dynamic knowledge graph organization and sleep-based memory consolidation.
Stanford & Harvard Paper Decodes Agentic AI's Demo-vs-Reality Gap
A new paper from Stanford, Harvard, UC Berkeley, and Caltech proposes a unified framework for understanding adaptation in Agentic AI systems, explaining why they often excel in demos but struggle in real-world applications.
Meta-Optimized Continual Adaptation for autonomous urban air mobility routing with ethical auditability baked in
This article details a framework, MOCA-E², for autonomous urban air mobility routing that achieves a 37% improvement in delivery time while incorporating ethical considerations and auditability.
QConAI: Balancing Probabilistic and Deterministic Systems for Reliable Agentic AI
Aaron Erickson at QCon AI NYC 2025 advocated for treating agentic AI as an engineering challenge, emphasizing the need for deterministic systems to constrain probabilistic models and improve reliability.
OpenAI’s Agent RFT: Reinforcement Fine-Tuning for Tool-Using Agents
OpenAI's Agent RFT, unveiled at QCon AI NYC 2025, uses reinforcement fine-tuning to improve tool-using agent performance by optimizing prompts and tasks before model adjustments.