邓力在威斯康星大学麦迪逊分校获得博士学位，而后在加拿大滑铁卢大学任教并获得终身正教授职位。随后，他以首席研究员的身份加入微软雷蒙德研究院，现任 2014年初成立的微软雷蒙德研究院深度学习技术中心研发部门负责人，同时出任微软人工智能首席科学家。邓力博士当前主要从事与大数据分析、自然语言文本、语义建模、语音、图像和多模式信号处理等关键业务应用有关的研发活动。除了拥有70多项专利和在领先的期刊和会议上发表超过300篇论文，邓力还是两本新书《深度学习：方法与应用》（NOW出版社，2014年）和《自动语音识别：深度学习方法》（Springer出版社，2015年）的作者。邓力是IEEE、美国声学学会和ISCA的院士，曾任 IEEE信号处理学会理事会理事，还出任过IEEE 信号处理杂志和《音频、语音与语言处理学报》主编。邓力博士在产业规模的深度学习和人工智能技术方面的工作影响了信息处理的各个领域。他的工作帮助重新点燃了（深度）神经网络在当前这个大数据、大计算时代的复苏，并且“因为在深度学习和自动语音识别方面的杰出贡献”得到多次表彰，包括 2013年IEEE SPS最佳论文奖和2015年IEEE SPS技术成就奖。

　　【演讲题目】人工智能：最新发展及未来挑战Artificial Intelligence: Recent Progress and Future Challenges

　　【摘要】Progress in AI research is rapidly having an impact in the marketplace. Basic AI technologies in machine learning, reasoning, and perception are leading to new capabilities in language translation, information extraction, computer vision, robot control (including self-driving cars), and scheduling and logistics. This talk will provide a brief tour of recent advances in AI technologies and some of their applications. It will then discuss three technical and engineering challenges of the future. (1) Unlike ordinary software, it is unclear when it is safe to trust AI systems. (2) Combined human-AI systems require good user interfaces in which the system can understand the human's intent and the human can predict the system's behavior. (3) Autonomous systems operating at high speed or in inaccessible locations cannot be monitored by humans. Is it possible to safely deploy them?

　　Thomas Dietterich ，AAAI前主席

　　PhD Stanford 1985 in Computer Science. Pioneer in machine learning. Inventor of error-correcting output codes, multiple-instance learning, and MAXQ hierarchical reinforcement learning. Current research focuses on robust AI and AI methods for computational sustainability and sustainable development.

　　【演讲题目】我们能从AI中期待什么? What Can We Expect from AI?

　　【摘要】AI (Artificial Intelligence) is becoming extremely hot, and there is not a single day on which one cannot see any news about AI. Indeed a number of big breakthroughs have been made in AI, such as self-driving cars, computer Go, speech recognition, image recognition, and machine translation, just to name a few. On the other hand, to build a machine that has the same level of intelligence as humans is also extremely difficult. What can and should we expect from AI in the foreseeable near future? In this talk, I will try to answer the question. I will argue that it is still difficult to implement “strong AI” under the current computer architecture. I will analyze the strength and limitation of machine learning which current “weak AI” heavily relies on, and point out that the level of intelligence of this weak AI, even with limitation, will still be continuously enhanced in the era of big data. I will then introduce how Huawei Noah’s Ark Lab is trying to revolutionize the telecommunication industry through innovations in AI. I will introduce our work on automatic control of flows on software defined networks using deep reinforcement learning, and intelligent assistance to engineers on network troubleshooting using deep learning.

　　李航，华为诺亚方舟实验室主任