报告时间：2017年9月21日（周四）上午9:00—11：00
报告地点：浙江大学紫金港校区蒙民伟楼402会议室
报告人：Ming Yin和Yexiang Xue　 博士
主持人：蔡登　教授

Title1: Peeking into the On-Demand Economyf

Abstract: Today, an increasing number of digital and mobile technologies have emerged to match customers, in almost real time, with a potentially global pool of self-employed labor, leading to the rise of the on-demand economy, which has brought about dramatic changes in our society. It creates new business models and new dynamics of labor allocation. It enables new models of computation, that is, human-in-the-loop computing. And it leads to new forms of knowledge creation—people all over the world are contributing to scientific studies in dozens of fields, either by making scientific observations as amateur scientists or by participating in online experiments as subjects.  Despite its already significant impacts, the on-demand economy has still been considered as a black-box approach to soliciting labor from a crowd of on-demand workers. Little is known about these workers and their aggregated behavior. In this talk, using the on-demand crowdsourcing platforms as an example, I present my attempts and findings on opening up this black box with a combination of experimental and computational approaches, with focuses on understanding who the on-demand workers are, how to model their unique working behavior, and how to improve their work experience.

Bio:Ming Yin is currently a postdoctoral researcher at Microsoft Research New York City. Starting in Fall 2018, she will join Purdue University as an Assistant Professor in the Department of Computer Science. Ming’s primary research interests lie in the interdisciplinary area of social computing and crowdsourcing. Her research has contributed to better understanding human behavior in social computing and crowdsourcing systems through large-scale online behavior experiments, as well as incorporating the empirical insights from the behavioral data into developing models, algorithms, and interfaces to facilitate the design towards better systems.
More broadly, her research connects to the fields of applied artificial intelligence and machine learning, computational social science, human-computer interaction and behavioral economics. Ming’s work is published in top venues like WWW, CHI, AAAI and IJCAI. Ming is named as a Siebel Scholar (Class of 2017), and she has received Best Paper Honorable Mention at the ACM Conference on Human Factors in Computing Systems (CHI’16). Ming completed her PhD at Harvard University under the supervision of Professor Yiling Chen, and she has interned at Microsoft Research (New York City Lab and New England Lab), PARC, and Xerox Research previously.

Title2: Combining Reasoning and Learning for Data Science and Decision Making: Integrating Concepts from AI, Sustainability, and Scientific Discovery

Abstract: Problems at the intersection of reasoning, optimization, and learning often involve multi-stage inference and are therefore highly intractable. I will introduce a novel computational framework, based on embeddings, to tackle multi-stage inference problems. As a first example, I present a novel way to encode the reward allocation problem for a two-stage organizer-agent game-theoretic framework as a single stage optimization problem. The encoding embeds an approximation of the agents’ decision-making process into the organizer’s problem. We apply this methodology to eBird, a well-established citizen-science program for collecting bird observations, as a game called Avicaching. Our AI-based reward allocation was shown highly effective, surpassing the expectations of the eBird organizers and bird conservation experts. As a second example, I present a novel constant approximation algorithm to solve the so-called Marginal Maximum-A-Posteriori (MMAP) problem for finding the optimal policy maximizing the expectation of a stochastic objective. To tackle this problem, I propose the embedding of its intractable counting subproblems as queries to NP-oracles subject to additional XOR constraints. As a result, the entire problem is encoded as a single NP-equivalent optimization. The approach outperforms state-of-the-art solvers based on variational inference as well as MCMC sampling on probabilistic inference benchmarks, deep learning applications, as well as on a novel decision-making application in network design for wildlife conservation. Lastly, I will talk about how a novel integration of reasoning and learning has led to the discovery of new solar light absorbers by solving a dimensionality reduction problem to characterize the crystal structures of metal oxide materials using X-ray diffraction data.
Bio: Yexiang Xue is a Ph.D. candidate in the Department of Computer Science at Cornell University, working with Professors Carla Gomes and Bart Selman. Upon graduation, he will join Purdue University as an assistant professor in computer science starting Fall 2018. His research aims at combining large-scale constraint-based reasoning and optimization with state-of-the-art machine learning techniques to enable intelligent agents to make optimal decisions in high-dimensional and uncertain real-world applications. More specifically, his research focuses on scalable and accurate probabilistic reasoning techniques, statistical modeling of data, and robust decision-making under uncertainty. Mr. Xue’s work is motivated by key problems across multiple scientific domains, including artificial intelligence, machine learning, renewable energy, materials science, citizen science, urban computing, and ecology, with an emphasis on developing cross-cutting computational methods for applications in the areas of computational sustainability and scientific discovery. Mr. Xue’s work received the Innovative Application Award at IAAI-17 and was featured as the cover article and the Editor’s Choice in the journal Combinatorial Science of the American Chemical Society.