My research focuses on AI for scientific discovery, with particular emphasis on applying modern machine learning models to materials discovery, chemical reaction modeling, and machine learning-assisted molecular dynamics. My goal is to develop efficient, data-driven approaches that can predict and design molecular and materials behavior, accelerating the discovery of novel materials with targeted properties and enabling high-impact applications such as carbon capture, battery and energy-storage materials, and other sustainability relevant technologies. Second, I develop deep learning methods for set-structured and relational data, which arise naturally across many scientific and machine learning settings. Many problems such as object-centric relation learning, feature and instance selection, and learning over unordered collections, can be cast within a set-based learning framework. By building principled and scalable algorithms for these structures, I aim to strengthen the foundations that support robust modeling in scientific machine learning and beyond.
Open to collaboration
Let’s explore ideas together.
Working on AI for scientific discovery or the foundations of machine learning? If you see an opportunity to collaborate, I’d love to hear from you.
Publications
* := Equal Contribution
MotionJEPA: Preventing Temporal Feature Collapse by Capturing Visual Changes in Latent Space
Markus Karmann*, Shile Li*, Christian Internò, Bruno Andreis, David Klindt, Randall Balestriero, Jindong Gu, Philip Torr, Qi Zhang, Peng-Tao Jiang, Hao Zhang, Bo Li, Onay Urfalioglu
arXiv preprint
arXiv 2026
Procedural Pretraining for Molecular Property Prediction
Moritz Friedemann, Zachary Shinnick, Philip Torr, Bruno Andreis
arXiv preprint
arXiv 2026
Online Surrogate Repair: Decoupling High-Fidelity Feedback from Search Length in Closed-Loop Discovery
Xiaotang Feng, Philip Torr, Bruno Andreis
arXiv preprint
arXiv 2026
LS-Merge: Merging Language Models in Latent Space
Bedionita Soro*, Aoxuan Silvia Zhang*, Bruno Andreis*, Jaehyeong Jo, Song Chong, Sung Ju Hwang
International Conference on Learning Representations
ICLR 2026
Robust Molecular Property Prediction via Densifying Scarce Labeled Data
Jina Kim*, Jeffrey Willette*, Bruno Andreis*, Sung Ju Hwang
ICML 2025 Generative AI and Biology (GenBio) Workshop
ICML 2025
Diffusion-based Neural Network Weights Generation
Bedionita Soro*, Bruno Andreis*, Hayeon Lee, Wonyong Jeong, Song Chong, Frank Hutter, Sung Ju Hwang
International Conference on Learning Representations
ICLR 2025
Instruction-Guided Autoregressive Neural Network Parameter Generation
Bedionita Soro*, Bruno Andreis*, Song Chong, Sung Ju Hwang
ICLR Workshop on Neural Network Weights as a New Data Modality 2025
ICLR 2025
Set-based Neural Network Encoding Without Weight Tying
Bruno Andreis, Soro Bedionita, Philip H.S. Torr, Sung Ju Hwang
Conference on Neural Information Processing Systems
NeurIPS 2024
Scalable Set Encoding with Universal Mini-Batch Consistency and Unbiased Full Set Gradient Approximation
Jeffrey Willette*, Seanie Lee*, Bruno Andreis, Kenji Kawaguchi, Juho Lee, Sung Ju Hwang
International Conference on Machine Learning
ICML 2023
Set-based Meta-Interpolation for Few-Task Meta-Learning
Seanie Lee*, Bruno Andreis*, Kenji Kawaguchi, Juho Lee, Sung Ju Hwang
Conference on Neural Information Processing Systems
NeurIPS 2022
Distortion-Aware Network Pruning and Feature Reuse for Real-time Video Segmentation
Hyunsu Rhee, Dongchan Min, Sunil Hwang, Bruno Andreis, Sung Ju Hwang
Machine Learning for Autonomous Driving Workshop
NeurIPS 2022
Set Based Stochastic Subsampling
Bruno Andreis, Seanie Lee, A. Tuan Nguyen, Juho Lee, Eunho Yang, Sung Ju Hwang
International Conference on Machine Learning
ICML 2022
Mini-Batch Consistent Slot Set Encoder for Scalable Set Encoding
Bruno Andreis, Jeffrey Willette, Juho Lee, Sung Ju Hwang
Conference on Neural Information Processing Systems
NeurIPS 2021
Dynamic Detection-Tracking Switching
Bruno Andreis, Junhyeon Park, Sung Ju Hwang, Minwoo Kim
Tenth International Conference on Ubiquitous and Future Networks
ICUFN 2018