課程名稱： AI Alignment
課程內容：
Abstract: The course introduces fundamental concepts in AI safety and alignment, providing an opportunity for participants to engage with, evaluate, and discuss these concepts. AI alignment aims to ensure that AI systems are developed in accordance with human values, goals, and preferences, with the goal of preventing harm, promoting fairness and transparency, and positively contributing to society while minimizing risks. However, achieving alignment poses challenges, including ensuring value alignment across diverse contexts, addressing conflicts between AI goals and human objectives, preventing unintended consequences, and navigating ethical considerations and long-term societal impacts. As AI systems become increasingly capable and general, questions remain about how to build systems that are controllable, aligned with human intentions, and interpretable. The course begins with an introduction to foundational technical topics such as deep learning, generative AI, large language models, and reinforcement learning. Building upon this technical foundation, the course covers AI safety considerations, including topics such as reinforcement learning for human feedback, scalable oversight mechanisms, mechanistic interpretability, technical governance approaches, and best practices for deploying and testing AI projects responsibly.
課程時間：2024.06.24—2024.06.28
課程地點：E111
課程主講人簡介：
陸海兵，圣克拉拉大學萊維商學院教授，系主任。陸教授的研究領域為人工智能公平與治理、數據安全與隱私、大數據分析和云計算，研究成果入選《聯合國報告》，并獲得2021年度AACSB創新激勵獎、2023年度M&SOM最佳論文獎，數十篇學術論文在IEEE Transactions on Dependable and Secure Computing, IEEE Transactions on Intelligent Transportation Systems, IEEE Transactions on Big Data, ACM Transactions on MIS, INFORMS Journal on Computing, M&SOM等國際高水平頂刊，多次入選包括S&P、KDD、ICDM和ICDE等計算機科學等國際頂級學術會議。陸教授學術和專業見解發表在《紐約時報》、《今日美國》、《福布斯》、《連線》等媒體雜志上，獲得PARC、Groove、GEIRI North America、Ultimate Software、AKTANA和Markkula Center等研究項目。
Haibing Lu,Professor and Department Co-Chair, Information Systems & Analytics，Leavey School of Business，Santa Clara University.
Dr. Lu's research expertise spans a wide range of areas, including AI fairness and governance, data security & privacy, big data analytics and cloud computing. His notable achievements include being featured in the UN Report, winning the prestigious AACSB Innovations That Inspires Award, receiving the M&SOM Best Paper award, and repeated recognition through the SCU Leavey Business School's honors for Exceptional Research, Teaching, and Service.
Dr. Lu's scholarly contributions comprise a collection of highly-cited papers published in renowned journals such as IEEE Transactions on Dependable and Secure Computing, IEEE Transactions on Intelligent Transportation Systems, IEEE Transactions on Big Data, ACM Transactions on MIS, INFORMS Journal on Computing, M&SOM, etc. His research has also been presented at prominent computer science conferences, including S&P, KDD, ICDM, and ICDE.
Dr. Lu's expertise and insights have been featured in The New York Times, USA Today, Forbes, WIRED, etc. His research initiatives have been supported by a number of institutions including PARC, Groove, GEIRI North America, Ultimate Software, AKTANA, and Markkula Center, among others.
