Symposium on Mathematical Foundations of Trustworthy Learning

12th - 16th October, 2025

Monte Verita Congress Center, Switzerland

In today's data-driven world, machine learning plays an increasingly pivotal role, but with this prominence comes the need for rigorous mathematical underpinnings to ensure trustworthiness. There has been a recent flurry of theoretical research activities that examine fundamental principles such as privacy, adversarial robustness, reproducibility, generalization, and other societal aspects of machine learning algorithms.

Our aim is to create a forum where researchers at all career stages can engage in a vibrant exchange of ideas, foster collaborations, identify pressing challenges, and collectively advance the mathematical foundations of trustworthy machine learning. This symposium will bring together a unique diverse mix of leading researchers from (i) different fields (such as theoretical computer science, statistics, optimization, causality), (ii) different parts of the world (in Europe, North America, and Asia), and (iii) a balanced group of junior and senior researchers. The planned session topics will include

● Theory of deep learning - stat. / generalization viewpoint
● Theory of deep learning - optimization viewpoint
● Privacy-aware machine learning
● Reproducible optimization and learning
● Robustness under distribution shifts and adversarial attacks, causality
● Societal foundations of machine learning

Invited Speakers

Avatar

Rediet Abebe
Ellis / Uni Tuebingen
Avatar

David M. Blei
Columbia University
Avatar

Cristina Butucea
ENSAE‑CREST
Avatar

Surbhi Goel
University of Pennsylvania
Avatar

Gautam Kamath
University of Waterloo
Avatar

Amin Karbasi
Yale University
Avatar

Hongseok Namkoong
Columbia University
Avatar

Sewoong Oh
University of Washington
Avatar

Jessica Sorrell
Johns Hopkins University

Programme (Tentative)

Afternoon Arrival
Evening Welcome Reception
9:00 – 12:00 Deep Learning Theory I
2 keynote talks + 2 contributed talks
15:00 – 18:00 Arrival & Private-aware Learning
2 keynote talks + 2 contributed talks
Evening Poster Session / Networking
9:00 – 12:00 Generalization and Robustness I
2 keynote talks + 2 contributed talks
15:00 – 18:00 Reproducible Learning
2 keynote talks + 2 contributed talks
Evening Panel Discussion
9:00 – 12:00 Deep Learning Theory II
2 keynote talks + 2 contributed talks
15:00 – 18:00 Excursion
Evening Poster Session / Networking
9:00 – 12:00 Generalization and Robustness II
2 keynote talks + 2 contributed talks
15:00 – 18:00 Societal Foundations of ML
2 keynote talks + 2 contributed talks
Evening Departure

Organizing Committee

Location

Monte Verita Congress Center

Strada Collina 84
CH-6612 Ascona

Registration

Registration Fee: 350 CHF per person.

Please note that registration is binding and cannot be canceled or refunded.

Registration will close on August 31st, 2025.

Registration Link


Dear Participant, please follow the two steps below to register for the conference and book your meals and accommodation at Monte Verità Centre.

Please note that payments can only be made via credit card, TWINT, Apple Pay, or Google Pay.

  1. Step 1 - registration and payment of the conference registration fee

    Fill out and submit the registration form:
    Link: https://www.bi.id.ethz.ch/csfweb/faces/anonymous/event/registerparticipant.xhtml?eventtype=generalevent&eventid=3l3o3e3z-fe7tfa-lny6q7mk-1-lpmyjtl6-62r

    Once you have successfully paid the registration fee, you will receive two emails:

    1. The first email will be sent by the Saferpay platform, confirming payment of the registration fee. (Please check your spam folder.)
    2. The second email, from the CSF platform, will contain information for the booking of your meals and the bedroom.

  2. Step 2 – booking and payment of the meals & bedroom fee

    Link: http://shop2.monteverita.org/en/

    Enter the event code 728084.

    Book your meals and bedroom latest 15 days prior to arrival. After that date the platform is closed.

    Meals:
    • Please note that all participants must book the meal package at Restaurant Monte Verità, even if staying at an external hotel.
    • In case you are not attending the whole conference, it is also possible to book the meals only for the duration of your stay.

    Bedroom:
    • The number of single rooms is limited, and some have a shared bathroom.
    • If you wish to share a twin room with a colleague, please provide both names when booking.
    • If you wish to book a shared double room with an accompanying person (not active participant), please provide both names when booking. Be aware that accompanying persons pay the standard room price and are not entitled the CHF 28.- additional price reduction per person and night granted by CSF/ETHZ to active participants.
    • Rooms can only be booked for the conference duration via this platform. If you wish to book a pre- and/or a post-night, please contact the Hotel Reception Monte Verità (info@monteverita.org).

    In case there is no availability or in case you do not wish to stay at Hotel Monte Verità, you can look for another hotel in the nearby Ascona village (15 to 20 min. walking distance from Monte Verita Centre).


Contact email: mml2025@ethz.ch

Sponsors

This symposium is generously supported by ETH Institute of Machine Learning, ETH Foundations of Data Science, ETH AI Center, Congressi Stefano Franscini, and BCAM.