Date: Tuesday, February 28, 2023
Title: Towards Autonomous Driving
Abstract: The automotive and transportation industry is going through a tectonic shift in the next decade with the advent of Connectivity, Automation, Sharing, and Electrification (CASE). Autonomous driving presents a historical opportunity to transform the academic, technological, and industrial landscape with advanced sensing and actuation, high definition mapping, new machine learning algorithms, smart planning and control, increasing computing powers, and new infrastructure with 5G, cloud and edge computing. Indeed, we have witnessed unprecedented innovation and activities in the past five years in R&D, investment, joint ventures, road tests and commercial trials, from auto makers, tier-ones, and new forces from the internet and high-tech industries.
In this talk, I will speak about this historical opportunity and challenges from technological, industrial and policy perspectives. I will address some of the core controversial and critical issues in the advancement of autonomous driving: open vs closed, Lidar vs cameras, progressive L2-L3-L4 vs new L4, autonomous capabilities vs V2X, Robotaxi vs vertical, China vs global, automakers vs new players, and the evolutional path and end game.
Bio: Dr. Ya-Qin Zhang is Chair Professor of AI Science at Tsinghua University, and Dean of Institute for AI Industry Research of Tsinghua University (AIR). He was the President of Baidu Inc. from 2014 to 2019. Prior to Baidu, Dr. Zhang was a Microsoft executive for 16 years with different key positions, including Managing Director of Microsoft Research Asia, Chairman of Microsoft China, and Corporate Vice President and Chairman of Microsoft Asia R&D.
Dr. Zhang was elected to the Chinese Academy of Engineering (CAE), the American Academy of Arts and Sciences (AAA&S), the Australian Academy of Technology and Engineering (ATSE), the National Academy of Inventors (NAI), and the Euro-Asia Academy of Sciences. He is a Fellow of IEEE and CAAI. He is one of the top scientists and technologists in digital video and AI, with over 500 papers, 60 granted US patents, and 11 books. His original research has become the basis for start-up ventures, new products, and international standards in digital video, cloud computing, and autonomous driving.
He serves on the Board of Directors of four public companies. He is on the industry board of United Nation Development Program (UNDP), and AI global council of the Davos World Economic Forum. He is the Chairman of world’s largest open autonomous driving platform “Apollo” alliance with over 200 global partners. He has been an active speaker in global forums including APEC, Davos, United Nations, and Bo’Ao Asia Forum.
Maarten de Rijke
University of Amsterdam
Date: Wednesday, March 1, 2023
Title: Beyond-Accuracy Goals, Again
Abstract: Improving the performance of information retrieval systems tends to be narrowly scoped. Often, better prediction performance is considered the only metric of improvement. As a result, work on improving information retrieval methods usually focuses on improving the methods’ accuracy. Such a focus is myopic. Instead, as researchers and practitioners we should adopt a richer perspective measuring the performance of information retrieval systems. I am not the first to make this point, but I want to highlight dimensions that broaden the scope considered so far and offer a number of examples to illustrate what this would mean for our research agendas.
First, trustworthiness is a prerequisite for people, organizations, and societies to use AI-based, and, especially, machine learning- based systems in general, and information retrieval systems in particular. Trust can be gained in an intrinsic manner by revealing the inner workings of an AI-based system, i.e., through explainability. Or it can be gained extrinsically by showing, in a principled or empirical manner, that a system upholds verifiable guarantees. Such guarantees should obtained for the following dimensions (at a minimum): (i) accuracy, including well-defined and explained contexts of usage; (ii) reliability, including exhibiting parity with respect to sensitive attributes; (iii) repeatable and reproducible results, including audit trails; (iv) resilience to adversarial examples, distributional shifts; and (v) safety, including privacy-preserving search and recommendation.
Second, in information retrieval, our experiments are mostly conducted in controlled laboratory environments. Extrapolating this information to evaluate the real-world effects often remains a challenge. This is particularly true when measuring the impact of information retrieval systems across broader scales, both temporally and spatially. Conducting controlled experimental trials for evaluating real-world impacts of information retrieval systems can result in depicting a snapshot situation, where systems are tailored towards that specific environment. As society is constantly changing, the requirements set for information retrieval systems are changing as well, resulting in short-term and long-term feedback loops with interactions between society and information retrieval systems.
Maarten de Rijke is a Distinguished University Professor of Artificial Intelligence and Information Retrieval at the University of Amsterdam. He is also the scientific director of the national Innovation Center for Artificial Intelligence (ICAI). His research is focused on designing trustworthy technology to connect people to information, particularly search engines, recommender systems, and conversational assistants. His work targets two key questions: How can we create intrinsic trust in information retrieval systems, that is, align their reasoning process with human expectations? And how can we establish extrinsic trust in information retrieval systems, that is, establish verifiable guarantees on their behavior?
Date: Thursday, March 2, 2023
Title: Learning to Understand Audio and Multimodal Content
Abstract: Music, podcasts and audiobooks are rich audio content types with deep listener engagement. Search and recommendation across these content types requires a deep understanding of their content, across audio, text and multimodal content. In this talk, I discuss some of the challenges and opportunities in understanding this content. This deep understanding of content enables us to delight our users and expand the reach of our content creators. As part of enabling wider academic research into podcast content understanding, Spotify Research has released a podcasts dataset with 120,000 hours of podcasts in English and Portuguese.
Dr. Rosie Jones is Director of Research at Spotify, where she leads a team of research scientists working on search and natural language understanding for music and podcasts. Before that, she worked at Microsoft, in the New England Research and Development Center (NERD), on conversational understanding. Prior to this she was Director of Data Science at MediaMath, where she led a team conducting research on large-scale machine learned models for display advertising. She moved to MediaMath via an acquisition from Akamai. Prior to working at Akamai, Dr Jones spent 8 years as a scientist at Yahoo! Labs, working on search and sponsored search. Her research interests include online user behavior, conversational understanding systems, web search, natural language processing and computational advertising. Dr. Jones has a PhD in Language and Information Technologies from Carnegie Mellon University and a B.Sc. from the University of Sydney, majoring in Computer Science. Her research at Microsoft, Akamai and Yahoo! has led to publications, product improvements and 14 issued patents, as well as academic service including senior PC member for COLING and SIGIR. Dr Jones is a Senior Member of the ACM.