iiinova — where your digital self keeps evolving

CSML IIT Lab @pontilgroup.bsky.social · 25/08/2026

We invite 4-page, non-archival short papers, with accepted work presented as posters and selected contributions invited for talks. 📅 Submission deadline: August 29 (AoE) 📍 Paris — NeurIPS 2026, December 12 or 13 🌐 Website: representations-physical-sciences.github.io/workshop-2026/

representations-physical-sciences.github.io

Representations for the Physical Sciences · NeurIPS 2026

Representations for the Physical Sciences — a NeurIPS 2026 workshop on self-supervision, transfer, sampling, and tokenization.

000

CSML IIT Lab @pontilgroup.bsky.social · 25/08/2026

Representations for the Physical Sciences Workshop @ NeurIPS 2026🇫🇷 We’re bringing together researchers working on AI for science, representation learning, and physical systems, with a particular focus on self-supervision, transfer learning, adaptive sampling, and tokenization.

120

CSML IIT Lab @pontilgroup.bsky.social · 08/07/2026

📍 Thursday, July 9 · 5:00–6:45 PM KST · Hall A · Poster #4517 Generalization of Gibbs and Langevin Monte Carlo Algorithms in the Interpolation Regime icml.cc/virtual/2026... If you're attending ICML, we'd be delighted to see you at the posters and discuss our work.

icml.cc

ICML Poster Generalization of Gibbs and Langevin Monte Carlo Algorithms in the Interpolation RegimeICML 2026

010

CSML IIT Lab @pontilgroup.bsky.social · 08/07/2026

📍 Wednesday, July 8 · 5:00–6:45 PM KST · Hall A · Poster #4401 Representation Learning for Equivariant Inference with Guarantees icml.cc/virtual/2026...

icml.cc

ICML Poster Representation Learning for Equivariant Inference with GuaranteesICML 2026

110

CSML IIT Lab @pontilgroup.bsky.social · 08/07/2026

📍 Wednesday, July 8 · 5:00–6:45 PM KST · Hall A · Poster #4113 Toward Scalable and Valid Conditional Independence Testing with Spectral Representations icml.cc/virtual/2026...

icml.cc

ICML Poster Toward Scalable and Valid Conditional Independence Testing with Spectral RepresentationsICML 2026

110

CSML IIT Lab @pontilgroup.bsky.social · 08/07/2026

ICML 2026 is underway in Seoul, and I'm delighted that our group has four papers on the program this week: 📍 Tuesday, July 7 · Hall A · Poster #4413 Outcome-Aware Spectral Feature Learning for Instrumental Variable Regression icml.cc/virtual/2026...

icml.cc

ICML Poster Outcome-Aware Spectral Feature Learning for Instrumental Variable RegressionICML 2026

121

CSML IIT Lab @pontilgroup.bsky.social · 17/12/2025

Almost 5 years in the making... "Hyperparameter Optimization in Machine Learning" is finally out! 📘 We designed this monograph to be self-contained, covering: Grid, Random & Quasi-random search, Bayesian & Multi-fidelity optimization, Gradient-based methods, Meta-learning. arxiv.org/abs/2410.22854

0139

CSML IIT Lab @pontilgroup.bsky.social · 28/11/2025

🚨 OpenReview might have leaked names, but it won't leak the best hyperparameters, unfortunately! 😅 Tired of the drama? Solve your HPO problems before the ICML deadline with this new monograph by our own Luca Franceschi & Massimiliano Pontil (& colleagues). arxiv.org/abs/2410.22854

arxiv.org

Hyperparameter Optimization in Machine Learning

Hyperparameters are configuration variables controlling the behavior of machine learning algorithms. They are ubiquitous in machine learning and artificial intelligence and the choice of their values ...

091

CSML IIT Lab @pontilgroup.bsky.social · 14/11/2025

He will also present an entropy-respecting forward–backward learning scheme that mitigates the inherent ill-posedness of stochastic learning problems. Join us for what promises to be a very insightful session!

040

CSML IIT Lab @pontilgroup.bsky.social · 14/11/2025

In this talk, Arthur Bizzi will introduce Neural Kolmogorov Equations, a deterministic and parallelizable framework for learning continuous-time stochastic processes using Forward and Backward Kolmogorov Equations.

110

CSML IIT Lab @pontilgroup.bsky.social · 14/11/2025

Abstract: Learning differential equations becomes substantially more challenging in the presence of stochasticity, as Neural SDEs typically require expensive, sequential integration during training.

110

CSML IIT Lab @pontilgroup.bsky.social · 14/11/2025

📢 Upcoming Talk at Our Lab We’re excited to host Arthur Bizzi from EPFL for a research talk next week! Title: Towards Neural Kolmogorov Equations: Parallelizable SDE Learning with Neural PDEs 🗓 Date: November 19 ⏰ Time: 16:00 CET 📍 Galileo Sala, CHT @iitalk.bsky.social

152

CSML IIT Lab @pontilgroup.bsky.social · 02/05/2025

Excited to share our group’s latest work at #AISTATS2025! 🎓 Tackling concentration in dependent data settings with empirical Bernstein bounds for Hilbert space-valued processes. 📍Catch the poster tomorrow! 🔁 See the original tweet for details!

030

CSML IIT Lab @pontilgroup.bsky.social · 09/04/2025

DeltaProduct is here! Achieve better state tracing through highly parallel execution. Explore more!🚀

041

CSML IIT Lab @pontilgroup.bsky.social · 15/01/2025

P11] (submitted to The Journal of Chemical Physics) chemrxiv.org/engage/chemr... Kooplearn library: kooplearn.readthedocs.io/latest/ For the longer version of the thread, you can take a look at this blog post: vladi-iit.github.io/posts/2024-1...

chemrxiv.org

Slow dynamical modes from static averages

In recent times, efforts are being made at describing the evolution of a complex system not through long trajectories, but via the study of probability distribution evolution. This more collective app...

020

CSML IIT Lab @pontilgroup.bsky.social · 15/01/2025

Publications: [P1] NeurIPS 2022 arxiv.org/abs/2205.14027 [P2] NeurIPS2023 arxiv.org/abs/2302.02004 [P3] ICML2024 arxiv.org/abs/2312.13426 [P4] NeurIPS2023 arxiv.org/abs/2306.04520 [P5] ICLR 2024 arxiv.org/abs/2307.09912 [P6] NeurIPS2024 arxiv.org/abs/2405.12940

arxiv.org

Learning Dynamical Systems via Koopman Operator Regression in Reproducing Kernel Hilbert Spaces

We study a class of dynamical systems modelled as Markov chains that admit an invariant distribution via the corresponding transfer, or Koopman, operator. While data-driven algorithms to reconstruct s...

120

CSML IIT Lab @pontilgroup.bsky.social · 15/01/2025

14/ Looking ahead, we’re excited to tackle new challenges: • Learning from partial observations • Modeling non-time-homogeneous dynamics • Expanding applications in neuroscience, genetics, and climate modeling Stay tuned for groundbreaking updates from our team! 🌍

120

CSML IIT Lab @pontilgroup.bsky.social · 15/01/2025

🙏 Collaborations with the Dynamic Legged Systems group led by Claudio Semini and the Atomistic Simulations group led by Michele Parrinello enriched our research, resulting in impactful works like [P9, P10] and [P7, P11].

110

CSML IIT Lab @pontilgroup.bsky.social · 15/01/2025

12/ This journey wouldn’t have been possible without the inspiring collaborations that shaped our work. 🌟 Special thanks to Karim Lounici from École Polytechnique, whose insights were a major driving force behind many projects.

110

CSML IIT Lab @pontilgroup.bsky.social · 15/01/2025

11/ One of our most exciting results: [P8] NeurIPS 2024 proposed Neural Conditional Probability (NCP) to efficiently learn conditional distributions. It simplifies uncertainty quantification and guarantees accuracy for nonlinear, high-dimensional data.

Predicting the quantiles for opening/closing of the Chignolin protein in the next simulation step

120

CSML IIT Lab @pontilgroup.bsky.social · 15/01/2025

10/ [P7] NeurIPS 2024 developed methods to discover slow dynamical modes in systems like molecular simulations. This is transformative for studying rare events and costly data acquisition scenarios in atomistic systems.

110

CSML IIT Lab @pontilgroup.bsky.social · 15/01/2025

9/ Addressing continuous dynamics: [P6] NeurIPS 2024 introduced a physics-informed framework for learning Infinitesimal Generators (IG) of stochastic systems, ensuring robust spectral estimation.

110

CSML IIT Lab @pontilgroup.bsky.social · 15/01/2025

8/ 🌟 Representation learning takes center stage in: [P5] ICLR 2024 We combined neural networks with operator theory via Deep Projection Networks (DPNets). This approach enhances robustness, scalability, and interpretability for dynamical systems.

110

CSML IIT Lab @pontilgroup.bsky.social · 15/01/2025

7/ 📈 Scaling up: [P4] NeurIPS 2023 introduced a Nyström sketching-based method to reduce computational costs from cubic to almost linear without sacrificing accuracy. Validated on massive datasets like molecular dynamics, see figure.

Free energy surface of Chignolin protein folding

110

CSML IIT Lab @pontilgroup.bsky.social · 15/01/2025

6/ [P3] ICML 2024 addressed a critical issue in TO-based modeling: reliable long-term predictions. Our Deflate-Learn-Inflate (DLI) paradigm ensures uniform error bounds, even for infinite time horizons. This method stabilized predictions in real-world tasks; see the figure.

Effects of metric distortion in learning eigenvalues (left) and stabilization of forecasting (right) for Ornstein-Uhlenbeck process

120

CSML IIT Lab @pontilgroup.bsky.social · 15/01/2025

5/ [P2] NeurIPS 2023 advanced TOs with theoretical guarantees for spectral decomposition—previously lacking finite sample guarantees. We developed sharp learning rates, enabling accurate, reliable models for long-term system behavior.

110

CSML IIT Lab @pontilgroup.bsky.social · 15/01/2025

4/ 🔑 The journey began with: [P1] NeurIPS 2022 We introduced the first ML formulation for learning TO, which led to the development of the open-source Kooplearn library. This step laid the groundwork for exploring the theoretical limits of operator learning from finite data.

120

CSML IIT Lab @pontilgroup.bsky.social · 15/01/2025

3/TOs describe system evolution over finite time intervals, while IGs capture instantaneous rates of change. Their spectral decomposition is key for identifying dominant modes and understanding long-term behavior in complex or stochastic systems.

110

CSML IIT Lab @pontilgroup.bsky.social · 15/01/2025

2/ 🌐 Our work revolves around Markov/Transfer Operators (TO) and their Infinitesimal Generators (IG)—tools that allow us to model complex dynamical systems by understanding their evolution in higher-dimensional spaces. Here’s why this matters.

110

CSML IIT Lab @pontilgroup.bsky.social · 15/01/2025

1/ 🚀 Over the past two years, our team, CSML, at IIT, has made significant strides in the data-driven modeling of dynamical systems. Curious about how we use advanced operator-based techniques to tackle real-world challenges? Let’s dive in! 🧵👇

153

CSML IIT Lab @pontilgroup.bsky.social · 15/01/2025

An inspiring dive into understanding dynamical processes through 'The Operator Way.' A fascinating approach made accessible for everyone—check it out! 👇👀

041

Reposted by CSML IIT Lab

Riccardo Grazzi @riccardograzzi.bsky.social · 10/12/2024

Excited to present "Unlocking State-Tracking in Linear RNNs Through Negative Eigenvalues" at the M3L workshop at #NeurIPS buff.ly/3BlcD4y If interested, you can attend the presentation the 14th at 15:00, pass at the afternoon poster session, or DM me to discuss :)

buff.ly

Unlocking State-Tracking in Linear RNNs Through Negative Eigenvalues

Linear Recurrent Neural Networks (LRNNs) such as Mamba, RWKV, GLA, mLSTM, and DeltaNet have emerged as efficient alternatives to Transformers in large language modeling, offering linear scaling with…

093

Reposted by CSML IIT Lab

Pietro Novelli @pienovelli.bsky.social · 12/12/2024

In his book “The Nature of Statistical Learning” V. Vapnik wrote: “When solving a given problem, try to avoid a more general problem as an intermediate step”

183

CSML IIT Lab @pontilgroup.bsky.social · 10/12/2024

Join us at our posters and talks to connect, share ideas, and explore collaborations. 🚀✨

030

CSML IIT Lab @pontilgroup.bsky.social · 10/12/2024

🔬 Fine-tuning Foundation Models for Molecular Dynamics: A Data-Efficient Approach with Random Features ✍️ @pienovelli.bsky.social, L. Bonati, P. Buigues, G. Meanti, L. Rosasco, M. Pontil | 📅ML4PS Workshop, Dec 15.

130

CSML IIT Lab @pontilgroup.bsky.social · 10/12/2024

🔗 Unlocking State-Tracking in Linear RNNs Through Negative Eigenvalues ✍️ R. Grazzi, J. Siems, J. Franke, A. Zela, F. Hutter, M. Pontil 📃https://arxiv.org/abs/2411.12537 | 📅 Oral @ M3L workshop, Dec 14, 15:00 - 15:15.

130

CSML IIT Lab @pontilgroup.bsky.social · 10/12/2024

🌊 Learning the Infinitesimal Generator of Stochastic Diffusion Processes ✍️V. Kostic, H. Halconruy, @tdevergne.bsky.social, K. Lounici, M. Pontil 📃https://arxiv.org/abs/2405.12940 | 📅 Poster #5410 Dec 13, 16:30 - 19:30.

130

CSML IIT Lab @pontilgroup.bsky.social · 10/12/2024

🧮 Generalization of Hamiltonian algorithms ✍️A. Maurer 📃https://arxiv.org/abs/2405.14469 | 📅 Poster #3706 Dec 13, 16:30 - 19:30.

130

CSML IIT Lab @pontilgroup.bsky.social · 10/12/2024

🔍 Neural Conditional Probability for Uncertainty Quantification ✍️V. Kostic, G. Pacreau, @giaturri.bsky.social, @pienovelli.bsky.social, K. Lounici, M. Pontil 📃https://arxiv.org/abs/2407.01171 | 📅 Poster #4007 Dec 13, 11:00 - 14:00.

130

CSML IIT Lab @pontilgroup.bsky.social · 10/12/2024

⚙️ From Biased to Unbiased Dynamics: An Infinitesimal Generator Approach ✍️ @tdevergne.bsky.social, V. Kostic, M. Parrinello, M. Pontil 📃https://arxiv.org/abs/2406.09028 | 📅 Poster #3806 Dec 12, 16:30 - 19:30.

130

CSML IIT Lab @pontilgroup.bsky.social · 10/12/2024

🌍 Operator World Models for Reinforcement Learning ✍️ @pienovelli.bsky.social, @marcopra.bsky.social, M. Pontil, C. Ciliberto 📃https://arxiv.org/abs/2406.19861 | 📅 Poster #6907 Dec 12, 16:30 - 19:30.

130

CSML IIT Lab @pontilgroup.bsky.social · 10/12/2024

At #NeurIPS2024 🇨🇦 our group will present 7 contributions! These span a diverse array of topics: from theoretical advances in stochastic processes and reinforcement learning to applications in molecular dynamics and uncertainty quantification.

162

Reposted by CSML IIT Lab

Marco Pratticò @marcopra.bsky.social · 26/11/2024

🎉 I am happy to share that I co-authored my first paper, “Operator World Models for Reinforcement Learning,” published at #NeurIPS2024! 🚀 Glad to present it in Vancouver 🇨🇦. See you there! #AI #ReinforcementLearning @pontilgroup.bsky.social arxiv.org/pdf/2406.19861

arxiv.org

2113