张长旺_AIAgentAI代理业界进展和发展趋势

AI Agent业界进展与发展趋势张长旺 @ OPPO研究院目录AI Agent技术简史01AI Agent落地难题03AI Agent业界进展02AI Agent发展趋势04AI Agent技术简史01Artificial Intelligence: A Modern Approach, Stuart Russell and Peter Norvig (2003).We define AI as the study of agents that receive percepts from the environment and perform actions. Each such agent implements a function that maps percept sequences to actions, and we cover different ways to represent these functions, such as production systems, reactive agents, real-time conditional planners, neural networks, and decision-theoretic systems.Agent技术简史：规则、强化学习、大模型3124通过反复试验学习，不断调整他们的政策以优化长期回报。Deep Q-network基于LMM的智能体ChatGPT+图片语音条件->动作专家系统基于LLM的智能体ChatGPT|AutoGPT强化学习Agent大多模态模型Agent规则Agent大语言模型Agent严重依赖人工扩展性差通用性差难训练通用性差问题：规划推理能力弱可靠性差专业领域任务效果差AI Agent技术栈系统层应用层模型层硬件层Agent应用大语言模型：LLM/LMM信息检索模型：Embedding/Rank应用开发框架：LangChain/AutoGen/LlamaIndex/DifyGPU/TPU/CPU/NPUAI Agent业界进展02AI产品榜：对话助手、AI搜索、AI图像是热门应用OpenAI AI Agent进展：实时多模态、工具生态、运行时推理优化2023.03.14对话能力全面提升GPT-42024.01.10支持处理图像GPT-4V2024.05.13实时处理