Martin Saveski

Assistant Professor, University of Washington

msaveski [AT] uw.edu

Bio

My research develops tools for analyzing large-scale social data, aiming to provide a better understanding of social structure and behaviors online while also impacting the design of digital social systems. My recent work has focused on reducing political polarization online [ICWSM'22a, ICWSM'22b, WSDM'21, WWW'18], improving the health of online conversations [WWW'21, CSCW'21], and causal inference in social systems [KDD'17 Biometrika'19]. My research often falls at the intersections of Social Networks, Machine Learning, and Causal Inference.

I'm an Assistant Professor in the Information School at the University of Washington. Before joining UW, I was a Postdoc at Stanford University, mentored by Johan Ugander. I received my Ph.D. from MIT in 2020 under the supervision of Deb Roy. Prior to coming to MIT, I spent one year in Paris and one year in Barcelona doing a M.Sc. in Data mining and Knowledge Management. I got my B.Sc. from Staffordshire University with First Class honors in Computer Science. Throughout my graduate studies, I spent several summers doing internships in industry, including Yahoo! Labs, Amazon, LinkedIn, and Facebook.

Publications

Most recent publications on Google Scholar.
‡ indicates equal contribution.

Embedding Societal Values into Social Media Algorithms

Michael Bernstein, Angèle Christin, Jeffrey Hancock, Tatsunori Hashimoto, Chenyan Jia, Michelle S. Lam, Nicole Meister, Nathaniel Persily, Tiziano Piccardi, Martin Saveski, Jeanne Tsai, Johan Ugander, Chunchen Xu

JOTS'23: Journal of Online Trust and Safety. 2023.

Paper

Counterfactual Evaluation of Peer-Review Assignment Policies

Martin Saveski, Steven Jecmen, Nihar B. Shah, Johan Ugander

NeurIPS'23: Advances in Neural Information Processing Systems (Spotlight). 2023.

Paper Code

Engaging Politically Diverse Audiences on Social Media

Martin Saveski, Doug Beeferman, David McClure, Deb Roy

ICWSM'22: International AAAI Conference on Web and Social Media. 2022.

Paper

Perspective-taking to Reduce Affective Polarization on Social Media

Martin Saveski‡, Nabeel Gillani‡, Ann Yuan, Prashanth Vijayaraghavan, Deb Roy

ICWSM'22: International AAAI Conference on Web and Social Media. 2022.

Paper Code Data

The Structure of Toxic Conversations on Twitter

Martin Saveski, Brandon Roy, Deb Roy

WWW'21: The Web Conference 2021

Paper Video Code Data

Social Catalysts: Characterizing People Who Spark Conversations Among Others

Martin Saveski, Farshad Kooti, Sylvia Morelli Vitousek, Carlos Diuk, Bryce Bartlett, Lada Adamic

CSCW'21: Conference on Computer-Supported Cooperative Work and Social Computing

Paper Code

Me, My Echo Chamber, and I: Introspection on Social Media Polarization

Nabeel Gillani, Ann Yuan, Martin Saveski, Soroush Vosoughi, Deb Roy

WWW'18: International Conference on the World Wide Web. 2018. (Honarable mention)

Paper Video

Detecting Network Effects: Randomizing Over Randomized Experiments

Martin Saveski‡, Jean Pouget-Abadie‡, Guillaume Saint-Jacques, Weitao Duan, Souvik Ghosh, Ya Xu, Edo Airoldi

KDD'17: International Conference on Knowledge Discovery and Data Mining. 2017. (Research Track)

Paper Slides Video

One-Pass Ranking Models for Low-Latency Product Recommendations

Antonino Freno, Martin Saveski, Rodolphe Jenatton, Cédric Archambeau

KDD'15: International Conference on Knowledge Discovery and Data Mining. 2015. (Industry Track)

Paper Slides Poster

Item Cold-Start Recommendations: Learning Local Collective Embeddings

Martin Saveski, Amin Mantrach.

RecSys'14, ACM Conference Series on Recommender Systems. 2014.

Paper Slides Code

Embedding Societal Values into Social Media Algorithms

Michael Bernstein, Angèle Christin, Jeffrey Hancock, Tatsunori Hashimoto, Chenyan Jia, Michelle S. Lam, Nicole Meister, Nathaniel Persily, Tiziano Piccardi, Martin Saveski, Jeanne Tsai, Johan Ugander, Chunchen Xu

JOTS'23: Journal of Online Trust and Safety. 2023.

Paper

Counterfactual Evaluation of Peer-Review Assignment Policies

Martin Saveski, Steven Jecmen, Nihar B. Shah, Johan Ugander

NeurIPS'23: Advances in Neural Information Processing Systems (Spotlight). 2023.

Paper Code

Engaging Politically Diverse Audiences on Social Media

Martin Saveski, Doug Beeferman, David McClure, Deb Roy

ICWSM'22: International AAAI Conference on Web and Social Media. 2022.

Paper

Perspective-taking to Reduce Affective Polarization on Social Media

Martin Saveski‡, Nabeel Gillani‡, Ann Yuan, Prashanth Vijayaraghavan, Deb Roy

ICWSM'22: International AAAI Conference on Web and Social Media. 2022.

Paper Code Data

The Structure of Toxic Conversations on Twitter

Martin Saveski, Brandon Roy, Deb Roy

WWW'21: The Web Conference 2021

Paper Video Code Data

Social Catalysts: Characterizing People Who Spark Conversations Among Others

Martin Saveski, Farshad Kooti, Sylvia Morelli Vitousek, Carlos Diuk, Bryce Bartlett, Lada Adamic

CSCW'21: Conference on Computer-Supported Cooperative Work and Social Computing

Paper Code

Balanced Influence Maximization in the Presence of Homophily

Md Sanzeed Anwar, Martin Saveski, Deb Roy

WSDM'21: International Conference on Web Search and Data Mining

Paper Code Data

Algorithmic and Human Prediction of Success in Human Collaboration From Visual Features

Martin Saveski‡, Edmond Awad‡, Iyad Rahwan, Manuel Cebrian

Nature Scientific Reports. 2021

Paper Code Data

Testing for Arbitrary Interference on Experimentation Platforms

Jean Pouget-Abadie, Guillaume Saint-Jacques‡, Martin Saveski‡, Weitao Duan, Souvik Ghosh, Ya Xu, Edo Airoldi

Biometrika. 2019.

Paper

Observational Causal Inference Using Network Information

Yan Leng, Martin Saveski, Alex ‘Sandy’ Pentland, Dean Eckles

NeurIPS'19, Workshop on Graph Representation Learning. 2019.

Me, My Echo Chamber, and I: Introspection on Social Media Polarization

Nabeel Gillani, Ann Yuan, Martin Saveski, Soroush Vosoughi, Deb Roy

WWW'18: International Conference on the World Wide Web. 2018. (Honarable mention)

Paper Video

Detecting Network Effects: Randomizing Over Randomized Experiments

Martin Saveski‡, Jean Pouget-Abadie‡, Guillaume Saint-Jacques, Weitao Duan, Souvik Ghosh, Ya Xu, Edo Airoldi

KDD'17: International Conference on Knowledge Discovery and Data Mining. 2017. (Research Track)

Paper Slides Video

Human Atlas: A Tool for Mapping Social Networks

Martin Saveski, Eric Chu, Soroush Vosoughi, Deb Roy

WWW'16: International Conference on the World Wide Web. 2016. (Demo)

Paper Video

Topic Modeling in Twitter: Aggregating Tweets by Conversations

David Alvarez-Melis‡, Martin Saveski‡

ICWSM'16: International AAAI Conference on Web and Social Media. 2016. (Short Paper)

Paper Poster

Tracking the Yak: An Empirical Study of Yik Yak

Martin Saveski, Sophie Chou, Deb Roy.

ICWSM'16: International AAAI Conference on Web and Social Media. 2016. (Short Paper)

Paper Poster

One-Pass Ranking Models for Low-Latency Product Recommendations

Antonino Freno, Martin Saveski, Rodolphe Jenatton, Cédric Archambeau

KDD'15: International Conference on Knowledge Discovery and Data Mining. 2015. (Industry Track)

Paper Slides Poster

Item Cold-Start Recommendations: Learning Local Collective Embeddings

Martin Saveski, Amin Mantrach.

RecSys'14, ACM Conference Series on Recommender Systems. 2014.

Paper Slides Code

The Geography of Online News Engagement

Martin Saveski, Daniele Quercia, Amin Mantrach

Socinfo'14: International Conference on Social Informatics. 2014.

Paper

Joint semi-supervised Learning of Hidden Conditional Random Fields and Hidden Markov Models

Yann Soullard, Martin Saveski, Thierry Artières

Pattern Recognition Letters. 2013.

Paper

Web Services for Stream Mining: A Stream-Based Active Learning Use Case

Martin Saveski, Miha Grčar

ECML'11, Workshop on Planning to Learn and Service-Oriented Knowledge Discovery. 2011.

Paper

Automatic Construction of Wordnets by Using Machine Translation and Language Modeling

Martin Saveski, Igor Trajkovski

In Proceedings of Seventh Language Technologies Conference. 2010.

Paper Slides

Development of an English-Macedonian Machine Readable Dictionary by Using Parallel Corpora

Martin Saveski, Igor Trajkovski

Proceedings of ICT Innovations Conference. 2010.

Paper

Vitæ

Full Resume in PDF.

Website Design

Since I made this website, many people have found this Jekyll template useful [ 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38 ]. You can find all the code needed to build your website in this GitHub repo. Feel free to use it.
If you end up using it, please link to here and drop me an email. I'm always happy to see people using it.
Martin Saveski
msaveski [AT] uw.edu