Alfonso Lobos

Data Scientist at MAIDAP, Microsoft, PhD Graduate

Biography

I’m currently a Data Scientist at Microsoft AI Development Acceleration Program (MAIDAP). I did my PhD in the University of California, Berkeley in Industrial Engineering and Operations Research (IEOR). My current growth interests are how to take machine learning and optimization models from theory to practice in large scale systems in a holistic manner. As research interests, I have been interested in optimization, in particular first order methods and online optimization, machine learning, and online advertising. Other areas which I find fascinating are e-commerce, economic problems, and robustness.

Interests

Optimization
Machine Learning
First Order Methods
Primal Dual Methods
Online Advertising

Education

PhD student in Industrial Engineering and Operations Research (IEOR) , 2021

University of California, Berkeley
Msc. in IEOR, 2016

University of California, Berkeley
Msc. in IEOR, 2015

Pontificia Universidad Catolica de Chile
Bachellor in Industrial Engineering With a Major In Mathematics , 2012

Pontificia Universidad Catolica de Chile

Recent Publications

Alfonso Lobos, Paul Grigas, Zheng Wen. Joint Online Learning and Decision-making via Dual Mirror Descent. ICML, 2021.

PDF Code

Alfonso Lobos, Paul Grigas, Zheng Wen, Kuang-Chih Lee. Optimal Bidding, Allocation, and Budget Spending for a Demand-Side Platform with Generic Auctions. 2021.

PDF Code

Paul Grigas, Alfonso Lobos, Nathan Vermeersch. Stochastic In-Face Frank-Wolfe Methods for Non-Convex Optimization and Sparse Neural Network Training. 2019.

PDF Code

Junwei Pan, Yizhi Mao, Alfonso Lobos, Yu Sun, Aaron Flores. Predicting different types of conversions with multi-task learning in online advertising. KDD, 2019.

PDF

Older Publications

Junwei Pan, Jian Xu, Alfonso Lobos, Wenliang Zhao, Shengjun Pan, Yu Sun, Quan Lu. Field-weighted Factorization Machines for Click-Through Rate Prediction in Display Advertising. WWW, 2018.

PDF

Alfonso Lobos, Paul Grigas, Zheng Wen, Kuang-chih Lee. Optimal Bidding, Allocation and Budget Spending for a Demand Side Platform Under Many Auction Types. AdKDD, 2018.

PDF

Paul Grigas, Alfonso Lobos, Zheng Wen, Kuang-chih Lee. Profit Maximization for Online Advertising Demand-Side Platform. AdKDD, 2017.

PDF

Alfonso Lobos, Jorge Vera. Intertemporal stochastic sawmill planning: Modeling and managerial insights. Computers and Industrial Engineering, 2016.

PDF

Amirreza M. Khaleghi, Dong Xu, Zhenrui Wang, Mingyang Li, Alfonso Lobos, Jian Liu, Young-JunSon. A DDDAMS-based planning and control framework for surveillance and crowd control via UAVs and UGVs. Expert Systems with Applications, 2013.

PDF

Amirreza M. Khaleghi, Dong Xu, Alfonso Lobos, Sara Minaeian, Young-Jun Son, Jian Liu. Agent-based hardware-in-the-loop simulation for UAV/UGV surveillance and crowd control system. WSC, 2013.

PDF

Teaching and Awards

Best student paper award at the 2018 AdKDD & TargetAd workshop of the conference Knowledge Data Discovery (KDD). Also obtained a travel grant from the OLX group.

Obtained the 2018 Outstanding Graduate Instructor Award, in other words, being in the 10% best rated teacher assistants in my department.

I have been a Graduate Student Instructor (Teaching Assistant) at Berkeley of:

IEOR 240: Optimization Analytics (Fall 2016, 2017).
IEOR 290: Modern Optimization for Statistical Learning (Spring 2017, 2018).

Contact

alobos@berkeley.edu