蛋播视频一区,无码鲁丝一区二区,精品 久久 五月天,国产老熟女,五月草草在线观看,中文日韩欧美,情色一区二区三区,欧美日韩亚洲激情在线,亚洲制服在线香蕉

信息管理與電子商務(wù)研究前沿系列

TITLE:

More Than Double Your Impact: An Empirical Study of Match Offers on Charitable Crowdfunding Platforms

ABSTRACT:

To promote charitable giving, donation-based crowdfunding platforms adopted match offers, whereby leadership donors commit to matching the contribution of other donors at a given rate. While match offers have great potential to improve fundraising performance, a lot remains unknown about how and when match offers work. Leveraging the data from a donation-based crowdfunding platform, our study seeks to understand (1) how the suppliers of funds (donors) evaluate charitable projects with and without match offers differently, (2) how these donors' preferences toward match offers vary with their donation experience, and (3) how the demanders of funds (fundraisers) react to the introduction of match offers. At an individual level, we find that, on average, donors derive a higher utility when contributing to charitable projects with match offers than without them. Specifically, warm-list donors (recently active donors) are three times more likely to contribute to matched projects than unmatched projects, while cold-list donors (dormant donors) are twice more likely to do so. However, new donors, who have no historical donation records on the platform, are more interested in unmatched projects. At a market level, we focus on the ratio of matched projects over all the projects and find that a 1% increase in the matched project ratio leads to a 1.34% increase in funds asked by demanders and a 0.854% increase in the funds supplied by donors. Finally, we demonstrate the robustness of our findings in a transactional analysis with fine-grained controls at the project level. Our work is one of the first studies that connect micro-level data patterns with macro-level market evidence to disentangle the impact of match offers systematically.

TIME:2021/08/09, 9:30-11:00a.m.(GMT+8)ZOOM:

ID: 83012704884

密碼：123456

SPEAKER:

Xue (Jane) Tan is an assistant professor in the Department of Operations and Decision Technologies, Kelley School of Business, Indiana University. She received her Ph.D. in business administration from the Foster School of Business, University of Washington. Her research interests include social network analysis, social media fundraising, online volunteerism, and electronic commerce. She has published in Information Systems Research and Management Information Systems Quarterly.

人工智能領(lǐng)域研究前沿系列

TITLE:

Getting stuff into and out of knowledge graphs: towards bridging the knowledge-language gap

ABSTRACT:

Knowledge graphs have attracted tremendous interest in both industry and academia, thanks to their modelling flexibility and versatility. How to get information into and out of knowledge graphs is key to their effective application. In this talk, I will present our recent research that contributes towards knowledge graph construction and question answering. Specifically, I will describe two lines of work: (1) relation extraction and (2) complex question answering, both of which utilise data-efficient meta-learning techniques.

TIME:2021/08/09, 13:00-14:30p.m.(GMT+8)ZOOM:

ID：83384344901

參會鏈接: https://us02web.zoom.us/j/83384344901

密碼：410417

SPEAKER:

Yuan-Fang Li received his Ph.D. in computer science from National University of Singapore in 2006. He is currently a Senior Lecturer at Department of Data Science and AI, Faculty of Information Technology, Monash University, Australia. His research interests lie at the intersection of knowledge graphs, representation learning, natural language processing, and ontology languages. His research work has been published at top AI, NLP and Semantic Web conferences including AAAI, IJCAI, ACL, COLING, EMNLP, and ISWC.

信息管理與電子商務(wù)研究前沿系列

TITLE:

Optimal Freemium Pricing and Ephemeral State-dependent Recommendations of Digital Content

ABSTRACT:

Digital content industries, such as e-book, movie, and music, are rapidly growing and projected to reach $367 billion by 2027. Managing digital content entails great challenges. This talk will cover two recent research projects, in collaboration with a leading e-book platform with $270 million annual revenues, on two essential platform strategies – freemium pricing and recommendation. Digital content platforms commonly enlist freemium pricing – offering initial content (e.g., the first few e-book chapters) for free in hope to monetize later content. In the first project, we identify the optimal charging points of a sample of e-books (i.e., charging at which chapter) via a large-scale randomized field experiment involving 1.3 million customers. We then conduct text analytics to uncover the relationship between each e-book’s optimal charging point and its content dynamics (such as sentiment culminations), hence automating the conventionally intuition-laden, labor intensive pricing decision for millions of unique products, and escalating the platform’s annual revenues by 50%. In the second project, we propose personalized, dynamic, ephemeral state-dependent recommendation strategies (fixation on a single type of content, or forage for diverse content). Such strategies present a stark contrast with the literature that has focused on recommendation strategies that assimilate with each consumer’s enduring preference, while overlooking the inter-temporal ephemeral state, hence potentially resulting in consumption fatigue and profit erosion. By leveraging a large-scale randomized field experiment and a series of statistical analyses, we demonstrate the flexibility and value of the ephemeral state-dependent recommendation strategies. For instance, a recommendation incongruent with a consumer’s ephemeral state (i.e., diversification recommendation to a fixation-state consumer, or assimilation to a forage-state consumer) attracts downloading, but not reading; whereas the opposite is true for a recommendation congruent with the ephemeral state. Both research projects offer valuable guidance to digital content industries.

TIME:

2021/08/10,9:30-11:00a.m.(GMT+8)

ZOOM:

ID: 81179021804

密碼：123456

SPEAKER:

Dr. Natasha Zhang Foutz studies entertainment marketing, digital media, and prosocial marketing with machine learning, econometric, statistical, and experimental methods. Her research has been published in books and leading journals, such as Journal of Marketing Research and Marketing Science. She serves as Area Editor for Journal of the Academy of Marketing Science and on the editorial review boards of various publication outlets. She teaches marketing analytics, entertainment marketing, marketing management, and marketing models at the undergraduate, MBA, EMBA, and PhD levels. She has received various research, teaching, and service awards, including Best Paper awards at multiple conferences, the Mallen Award for lifetime published scholarly contributions to motion picture industry studies, the Management Science Meritorious Service Award for Best Reviewer of the Year, and the University of Virginia All-University Teaching Award.

運籌優(yōu)化與運營管理研究前沿系列

TITLE:

A Riemannian Block Coordinate Descent Method for Computing the Projection Robust Wasserstein Distance

ABSTRACT:

The Wasserstein distance has become increasingly important in machine learning and deep learning. Despite its popularity, the Wasserstein distance is hard to approximate because of the curse of dimensionality. A recently proposed approach to alleviate the curse of dimensionality is to project the sampled data from the high dimensional probability distribution onto a lower-dimensional subspace, and then compute the Wasserstein distance between the projected data. However, this approach requires to solve a max-min problem over the Stiefel manifold, which is very challenging in practice. In this talk, we propose a Riemannian block coordinate descent (RBCD) method to solve this problem. We analyze the complexity of arithmetic operations for RBCD to obtain an $\epsilon$-stationary point, and show that it significantly improves the corresponding complexity of existing methods. Numerical results on both synthetic and real datasets demonstrate that our method is more efficient than existing methods, especially when the number of sampled data is very large.

TIME:2021/08/11, 13:00-15:00p.m.(GMT+8)Tencent Meeting:

ID：430866385

密碼：123456

SPEAKER:

Shiqian Ma is currently a tenured associate professor in the Department of Mathematics at University of California, Davis. He received his BS in Mathematics from Peking University in 2003, MS in Computational Mathematics from the Chinese Academy of Sciences in 2006, and PhD in Industrial Engineering and Operations Research from Columbia University in 2011. Shiqian was an NSF postdoctoral fellow in the Institute for Mathematics and its Applications at the University of Minnesota during 2011-2012 and an assistant professor in the Department of Systems Engineering and Engineering Management at the Chinese University of Hong Kong during 2012-2017. His current research interests include theory and algorithms for large-scale optimization, and their various applications in machine learning, signal processing, statistics and bioinformatics. Shiqian received the INFORMS Optimization Society Best Student Paper Prize in 2010, and an honorable mention in the INFORMS George Nicholson Student Paper Competition in 2011. He was one of the finalists for the 2011 IBM Herman Goldstine fellowship. He received the Journal of the Operations Research Society of China Excellent Paper Award in 2016. Shiqian currently serves on the editorial board of Journal of Scientific Computing.

人工智能領(lǐng)域研究前沿系列

TITLE:

Ultra-wide Neural Networks and Neural Tangent Kernel

ABSTRACT:

I will talk about the result on the equivalence between the over-parameterized neural network and a new kernel, Neural Tangent Kernel. This equivalence implies two surprising phenomena: 1) the simple algorithm gradient descent provably finds the global optimum of the highly non-convex empirical risk, and 2) the learned neural network generalizes well despite being highly over-parameterized. I will also present empirical results showing Neural Tangent Kernel is a strong predictor.

TIME:2021/08/12, 10:00-12:00a.m.(GMT+8)ZOOM:

ID: 81320173222

參會鏈接: https://us02web.zoom.us/j/81320173222

密碼：159937

SPEAKER:

Simon S. Du is an assistant professor in the Paul G. Allen School of Computer Science & Engineering at University of Washington. His research interests are broadly in machine learning such as deep learning, representation learning and reinforcement learning. Prior to starting as faculty, he was a postdoc at Institute for Advanced Study of Princeton. He completed his Ph.D. in Machine Learning at Carnegie Mellon University. Previously, he studied EECS and EMS at UC Berkeley. He has also spent time at Simons Institute and research labs of Facebook, Google and Microsoft.

運籌優(yōu)化與運營管理研究前沿系列

TITLE:

Machine Learning in Strategic Environments

ABSTRACT:

In today’s increasingly connected world, it is rare that an algorithm will act alone. When a machine learning algorithm is used to make predictions or decisions about others who have their own preferences over the learning outcomes, it is widely observed that gaming behaviors may arise. This talk will discuss a general framework to address learning under potential gaming manipulation. We then illustrate the power of this framework with two vignettes: one addresses test time manipulation in offline learning whereas another addresses training time manipulation in online learning.

TIME:2021/08/13, 9:00-11:00a.m.(GMT+8)Tecent Meeting:

ID：987113142

密碼：123456

SPEAKER:

Haifeng Xu is the Alan Batson Assistant Professor in Computer Science at the University of Virginia. He works broadly on game theory and machine learning, with a particular focus on designing intelligent algorithms to efficiently learn and act in informationally complex settings. He publishes extensively in leading theoretical CS and machine learning conferences such as STOC, SODA, EC, NeurIPS and ICML. Prior to UVA, Haifeng was a postdoc at Harvard and obtained his PhD in Computer Science from the University of Southern California. His research has been recognized by multiple awards, including a Google Faculty Research Award, honorable mention for the ACM SIGecom Dissertation Award, runner-up for the IFAAMAS Victor Lesser Distinguished Dissertation Award, a Google PhD fellowship, the 2016 AAMAS best student paper award, and the 2016 SecMas Workshop best paper award.

上財信息

講座信息將提前預(yù)告

歡迎校內(nèi)外師生校友旁聽！