Rule-Based Reasoning on AIBussin โ€” AI applications, systems and bookshttps://aibussin.com/tags/rule-based-reasoning/Recent content in Rule-Based Reasoning on AIBussin โ€” AI applications, systems and booksHugoen-USThu, 12 Jun 2025 09:23:46 +0100Learning to Learn: A LATS-Based Framework for Self-Aware AI Pipelineshttps://aibussin.com/post/lats/Thu, 12 Jun 2025 09:23:46 +0100https://aibussin.com/post/lats/<h2 id="-summary">๐Ÿ“– Summary</h2> <p>In this post, we introduce the LATSAgent, an implementation of <a href="https://arxiv.org/pdf/2310.04406" target="_blank" class="paper-badge" style="display: inline-block; padding: 6px 10px; background: #f3f4f6; border-left: 4px solid #3b82f6; border-radius: 4px; margin: 4px 0; text-decoration: none; color: #1f2937;"> <strong>LATS</strong>: Language Agent Tree Search Unifies Reasoning.. </a> within the <a href="https://github.com/ernanhughes/co-ai">stephanie</a> framework. Unlike prior agents that followed a single reasoning chain, this agent explores multiple reasoning paths in parallel, evaluates them using multidimensional scoring, and learns symbolic refinements over time. This is our most complete integration yet of search, simulation, scoring, and symbolic tuning bringing together all of our previous work on sharpening, pipeline reflection, and symbolic rules into a unified, intelligent reasoning loop.</p>Programming Intelligence: Using Symbolic Rules to Steer and Evolve AIhttps://aibussin.com/post/symbolic/Wed, 04 Jun 2025 20:57:20 +0100https://aibussin.com/post/symbolic/<h2 id="-summary">๐Ÿงช Summary</h2> <p>&ldquo;What if AI systems could learn how to improve themselves not just at the level of weights or prompts, but at the level of strategy itself? In this post, we show how to build such a system, powered by symbolic rules and reflection.</p> <p>The paper <a href="https://arxiv.org/pdf/2406.18532v1" target="_blank" class="paper-badge" style="display: inline-block; padding: 6px 10px; background: #f3f4f6; border-left: 4px solid #3b82f6; border-radius: 4px; margin: 4px 0; text-decoration: none; color: #1f2937;"> <strong>Symbolic Agents</strong>: Symbolic Learning Enables Self-Evolving Agents </a> introduces a framework where <strong>symbolic rules</strong> guide, evaluate, and evolve agent behavior.</p>