Agenda

Dr. Gabriele Tolomei,  Assistant Professor of the University of Padua, Italy.

Machinelearned models are often described as black boxes. In many realworld applications however, models may have to sacrifice predictive power in favour of humaninterpretability. When this is the case, feature engineering becomes a crucial task, which requires significant and timeconsuming human effort. Whilst some features are inherently static, representing properties that cannot be influenced e.g., the age of an individual, others capture characteristics that could be adjusted e.g., the daily amount of carbohydrates taken. Nonetheless, once a model is learned from the data, each prediction it makes on new instances is irreversible assuming every instance to be a static point located in the chosen feature space. There are many circumstances however where it is important to understand i why a model outputs a certain prediction on a given instance, ii which adjustable features of that instance should be modified, and finally iii how to alter such a prediction when the mutated instance is input back to the model.In this talk, I present a technique that exploits the internals of a treebased ensemble classifier to offer recommendations for transforming true negative instances into positively predicted ones. I demonstrate the validity of this approach using a realworld online advertising application. First, I provide the audience with some background knowledge on the online advertising domain and define the specific business problem which this research work originates from and aims to solve. Concretely, I discuss the design and development of a Random Forest classifier that effectively separates between two types of ads low negative and high positive quality ads instances. Then, I introduce an algorithm that provides recommendations that aim to transform a low quality ad negative instance into a high quality one positive instance. Finally, I present the main results and findings obtained when evaluating this approach on a subset of the active inventory of a large ad network, Yahoo Gemini.

Bio sketch

Gabriele Tolomei joined the Department of Mathematics of the University of Padua, Italy, as Assistant Professor in July 2017. Since June 2014, he has been a Research Scientist at Yahoo Research in London, UK, where he is still jointly working as Academic Collaborator.He received his Ph.D. in Computer Science from Ca Foscari University of Venice, Italy in November 2011. Afterwards, he was a PostDoc Researcher at the same University, and he also collaborated as Research Associate with the High Performance Computing Lab of the ISTICNR in Pisa, Italy. His main research interests include Web Search, Machine Learning, and Computational Advertising. On those topics, he authored around 30 papers, which appeared in topmost international peerreviewed journals and conferences. Moreover, he filed 4 patents with the US Patent and Trademark Office.He has been serving as Program Committee member and reviewer of many relevant international conferences, including ICDM, ICDE, KDD, WWW, WSDM, SIGIR, CIKM, and IJCAI, and as reviewer of international journals like TKDE, TOIS, and TWEB. He is Professional Member of the ACM.