Sara Magliacane

Sara Magliacane

Assistant Professor

University of Amsterdam

Biography

I am an assistant professor in the Amsterdam Machine Learning Lab at the University Amsterdam and an ELLIS Scholar in the Interactive Learning and Interventional Representations program. During Spring 2022, I was visiting the Simons Institute in Berkeley for a semester on Causality.

The goal of my research is to find how can causality improve current machine learning (ML) algorithms, especially in terms of robustness, generalization across domains/tasks, and safety. My research focuses on three directions: causal representation learning (i.e. learning causal factors from high-dimensional data, e.g. sequences of images [1, 2, 3]), causal discovery (i.e. learning causal relations from data), and causality-inspired ML, e.g. how can ideas from causality help ML/RL adapt to new domains, nonstationarity and varying number of objects with different latent parameters, even when we cannot guarantee that we identified the true causal factors.

Previously I was a Research Scientist at MIT-IBM Watson AI lab and a postdoctoral researcher at IBM Research NY, working on methods to design experiments that would allow one to learn causal relations in a sample-efficient and intervention-efficient way. I received a PhD at VU Amsterdam on learning causal relations jointly from different experimental settings, even with latent confounders and small samples. During my PhD, I interned at Google Zürich and NYC. Previously, I studied Computer Engineering at Politecnico di Milano and Torino and at the University of Trieste.

Download my resumé .

Interests
  • Causal Representation Learning
  • Causal discovery
  • Causality-inspired ML and RL
  • Causality in general
  • Neurosymbolic/StarAI approaches
Education
  • PhD in Artificial Intelligence, 2017

    VU Amsterdam

  • MSc in Computer Engineering, 2011

    Politecnico di Milano, Politecnico di Torino (double degree)

  • BSc in Computer Engineering, 2008

    Università degli Studi di Trieste

Recent Publications

(2023). BISCUIT: Causal Representation Learning from Binary Interactions. UAI 2023.

PDF Cite Code Project Poster Slides URL

(2023). Causal Representation Learning for Instantaneous and Temporal Effects in Interactive Systems. ICLR 2023.

PDF Cite Code Poster URL

(2023). Graph Switching Dynamical Systems in Interactive Systems. ICML 2023.

PDF Cite Code Poster URL

(2023). Modulated Neural ODEs. NeurIPS 2023.

PDF Cite URL

Team

Master students

Avatar

Mara Pislar

Master student (UvA)

Avatar

Pieter Bouwman

Master student (UvA)

PhD students

Avatar

Daan Roos

PhD student (UvA)

Avatar

Danru Xu

PhD student (UvA)

Avatar

Jakub Řeha

PhD student (UvA)

Avatar

Mátyás Schubert

PhD student (UvA)

Avatar

Nadja Rutsch

PhD student (AUMC)

Avatar

Phillip Lippe

ELLIS PhD student (UvA/Qualcomm)

Avatar

Riccardo Massidda

ELLIS PhD student (University of Pisa/UvA)

Avatar

Roel Hulsman

PhD student (UvA)

Avatar

Yongtuo Liu

PhD student (UvA)

Close collaborators

Avatar

Fan Feng

PhD student (City University Hong Kong)

Guest researchers

Avatar

Aoqi Zuo

PhD student (University of Melbourne)

Alumni

Avatar

Andrea Conte

Master student (University of Torino)

Avatar

Angelos Nalmpantis

Master student (UvA)

Avatar

Danilo de Goede

Master student (UvA)

Avatar

Davide Talon

PhD student (IIT Genoa)

Avatar

Eva Sevenster

Master student (UvA)

Avatar

Frank Brongers

Master student (UvA)

Avatar

Ilze Amanda Auzina

PhD student at University of Amsterdam

Avatar

Willemijn de Hoop

Master student (UvA)

Avatar

Yohan Runhaar

Master student (UvA, Adyen)

News

News

Contact

For teaching matters, you can contact me via Canvas.

PhD students If you have questions about joining my group or AMLab, check this list first.

Other jobs Currently I don’t have any open vacancy for interns, research assistants or postdocs, but if I do, they will be announced on the UvA vacancies website.

Master students If you are a Master in AI student at the University of Amsterdam and are interested in causality, feel free to contact me for potential thesis topics. As a rule, I also don’t supervise Bachelor or Master students from other universities.