【星空正规app】第278期:The Future of AI: The Era of Experience and the Age of Design

2025-11-30 16:00:00-17:30:00
陈瑞球楼100号

人工智能正在迈向全新的“体验时代”。从早期的复杂游戏(如围棋、Dota2)到近期的大模型推理中取得里程碑式的突破,强化学习已经证明了其作为实现通用智能核心路径之一的巨大潜力。它标志着AI研究范式的根本性转变——从依赖静态、被动的数据集进学习,转向让智能体在动态、交互的“体验”中通过试错进行规划与决策。本次讲座将探讨先进强化学习的前沿思想与技术挑战,探讨如何突破范式,实现通用人工智能的宏伟目标。
Artificial intelligence is entering a new "era of experience." From early complex games like Go and Dota 2 to recent milestone breakthroughs in reasoning with large models, reinforcement learning has demonstrated its immense potential as one of the core pathways toward achieving general intelligence. It signifies a fundamental shift in AI research paradigms—moving from learning reliant on static, passive datasets to enabling intelligent agents to plan and make decisions through trial and error in dynamic, interactive "experiences." This lecture will explore the cutting-edge ideas and technical challenges in advanced reinforcement learning, delving into how to break through existing paradigms and achieve the grand goal of artificial general intelligence.

嘉宾介绍

Richard Sutton

2024年图灵奖获得者(Turing Award Laureate,2024)
演讲主题:The Future of AI: The Era of Experience and the Age of Design
Richard Sutton is a recipient of the ACM Turing Award, the highest distinction in computing science. He is also a fellow of the Royal Society of London, the Royal Society of Canada, and the Association for the Advancement of Artificial Intelligence. At the University of Alberta and Keen Technologies, he teaches and designs learning algorithms for artificial intelligence and reinforcement learning. At the Alberta Machine Intelligence Institute, he is Chief Scientific Advisor and a Canada CIFAR AI chair. Earlier, he studied at Stanford University and the University of Massachusetts, and worked at GTE Labs, AT&T Labs, and DeepMind. His scientific publications have been cited about 170,000 times and include the defining textbook on reinforcement learning. Sutton’s research emphasizes learning from the agent’s first-person experience without special support from a teacher or a prepared dataset. He is also a libertarian, a chess player, and a cancer survivor.
Richard Sutton是计算机科学最高荣誉——ACM图灵奖获得者,同时兼任伦敦皇家学会、加拿大皇家学会及国际人工智能促进协会会士。他在阿尔伯塔大学与Keen Technologies从事人工智能与强化学习算法的教学与设计工作,并担任阿尔伯塔机器智能研究所首席科学顾问及加拿大CIFAR人工智能讲席教授。早年曾就读于斯坦福大学与马萨诸塞大学,先后任职于GTE实验室、AT&T实验室和DeepMind公司。其学术论著被引用约17万次,其中包含强化学习领域的奠基性教材。萨顿的研究强调智能体应从第一人称视角的经验中自主学习,无需教师指导或预设数据集。他同时也是自由意志主义者、国际象棋爱好者和抗癌成功者。
Baidu
map