News

Aug 01, 2026 Our proposal for AXIOM: “Foundations of Efficient Deep Learning”, has been accepted to NeurIPS 2026 (Paris). Workshop: December 12, 2026. We bring together theory and practice to focus on efficientML and develop predictive principles for efficient learning.
Apr 12, 2026 PAPER 📝 We challenge the lottery ticket metaphor for understanding overparameterization. We propose Escape Dimensions as a substitute intuitive mental picture based on loss landscape theory. Check out the full article in the blog page.
Dec 15, 2025 TALK 🎤 Talk about our Flat Channels to Infinity for the ELLIS - Mathematics & Efficiency of Deep Learning reading group – link to video.
Dec 09, 2025 PAPER 📝 Our work on in-memory computing with memristive devices is out in Nature Machine Intelligence! We push the boundaries of what’s computable in-memory, delegating both inference, error computation and error updates on the hardware side to minimize slow and inefficient cross-talk with a generic CPU. Have a look at Actor–critic networks with analogue memristors mimicking reward-based learning
Dec 01, 2025 POSTER 🏞️ In San Diego for NeurIPS ‘25, presenting our Flat Channels to Infinity, Degeneracy in RNNs and Augmentation Techniques for Expand-and-Cluster.
Nov 25, 2025 PAPER 📝 Our extension of Expand-and-Cluster for overparameterized networks have been accepted at the NeurIPS UniReps workshop. We come up with ways to augment data specifically to extract informative signals from black-box networks. Check out Data Augmentation Techniques to Reverse-Engineer Neural Network Weights from Input-Output Queries.
Oct 01, 2025 POSTER 🏞️ In Frankfurt for two posters on RNN solution degeneracy and Toy models of identifiability for neuroscience at the Bernstein conference.
Sep 15, 2025 PAPER 📝 Measuring and Controlling Solution Degeneracy across Task-Trained Recurrent Neural Networks has been accepted to NeurIPS as spotlight!
Sep 15, 2025 PAPER 📝 Flat Channels to Infinity in Neural Loss Landscapes has been accepted to NeurIPS for a poster!
Sep 03, 2025 TALK 🎤 Gave a talk about the channels to infinity in loss landscapes in the Ploutos platform.
Aug 03, 2025 🌎 Currently in Woods Hole, MA (US) for the MIT Brain, Minds and Machines summer school.
Jun 17, 2025 PAPER 📝 We discover channels of slowly decreasing loss in network loss landscapes that lead to minima at infinite parameter norm. In the limit, these solutions implement Gated Linear Units using standard neurons. These channels are parallel to lines of saddle points generated by permutation symmetries. Read the paper: Flat Channels to Infinity in Neural Loss Landscapes.
Jun 05, 2025 POSTER 🏞️ Presented a poster at Frontiers in NeuroAI, Boston, on optimization challenges to recover connectivity from activity.
May 28, 2025 PAPER 📝 How degenerate is the solution space of identical RNNs trained on the same task? Check out our new paper on measuring and controlling degeneracy of RNNs. Fun collaboration with the Rajan Lab.
Mar 26, 2025 POSTER 🏞️ Presented a poster at Kempner Institute’s Spring Into Science meeting.
Mar 01, 2025 🌎 I am currently based in Boston for a 4 months visit in Kanaka Rajan’s lab. Reach out if you want to have a chat!
Sep 02, 2024 TALK 🎤 I gave a talk on Expand-and-Cluster for the EfficientML reading group – link to video.
Jul 21, 2024 TALK 🎤 POSTER 🏞️ In Vienna for ICML ‘24 and a visit to Tim Vogel’s lab in IST, Austria.
May 20, 2024 POSTER 🏞️ In Trieste for the Youth in High Dimensions meeting.
May 02, 2024 PAPER 📝 Expand-and-Cluster is accepted to ICML 2024!

🔍 We identify the weights of a neural network from simple input-output queries.
⚠️ Symmetries and overparameterised loss landscapes are heavily involved.
Feb 08, 2024 TALK 🎤 Gave a talk and visited the laboratory of Jakob Macke at the Tübingen AI Center, Germany.
Jan 31, 2024 TALK 🎤 Gave a talk at the annual Swiss Computational Neuroscience meeting.
Sep 26, 2023 POSTER 🏞️ In Berlin for a poster at the Bernstein conference.
Sep 01, 2023 TALK 🎤 Best presentation award at the NeuroLeman meeting 🎖️
Jun 06, 2023 TALK 🎤 Gave a talk and visited the group of Angelika Steger and Joao Sacramento in ETH, Zürich.