Ph.D. in Computer Science, 2023-2027,
EPFL (CLAIRE and LIAC Labs)
Advisors: Caglar Gulcehre and Philippe Schwaller
M.Sc. in Computer Science, 2020-2023
Mila and University of Montreal
B.Sc. in Electrical Engineering, 2015-2020
Sharif University of Technology
I'm currently thinking about ways to improve reasoning and alignment in service of scientific discovery in various domains such as math, chemistry, etc. This spans research questions across multiple axes such as search and planning to explore the vast range of ideas, tool-augmented agentic systems mitigating hallucinations, as well as measuring model's epistemic uncertainty of its own knowledge.
Prior to that, I worked on length and compositional generalization of foundation models, and even earlier, I worked on representation learning for out-of-distribution generalization in pursuit of system-2 reasoning (back in 2021-22)
Usually I can be found in my office on EPFL campus (BC 201). When not working, I might be found in MXD building practicing piano, and if not there, I'm very likely wandering in Swiss Alps or sailing in lac Léman.
Google Scholar Github Linkedin Twitter CV: Please send an email Last updated: Summer 2025
[For Recruiters and Industry Researchers]
I am actively looking for opportunities to get involved with industry research on reasoning, alignment, and AI for scientific discovery. If my profile is a fit for an opening in your team, I'd love to learn about the opportunity!
[For Students]
First of all, I truly appreciate your interest in working together. I am always happy to discuss new ideas, learn from each other, and share our findings! Please feel free to reach out. However, I can (time-permitting) only commit to projects that are feasible with moderate compute already accessible to you, as computing resources at EPFL and Mila cannot be shared externally.
I always strive to be a better person, be it as a researcher, a colleague, or a friend. I always welcome and appreciate constructive feedback. So this form is simply an invitation to share anything that could help me improve.
Please refer to google scholar for an up-to-date list of publications.
Algorithm Discovery With LLMs: Evolutionary Search Meets Reinforcement Learning
Anja Surina, Amin Mansouri, Lars Quaedvlieg, Amal Seddas, Maryna Viazovska, Emmanuel Abbe, Caglar Gulcehre
COLM 2025
Toward Causal Representation and Structure Learning
Amin Mansouri, MSc Thesis (Dean's Honor List)
On Provable Length and Compositional Generalization
Kartik Ahuja, Amin Mansouri
Preprint, 2024
Symbolic Autoencoding for Self-Supervised Sequence Learning
Mohammad Hossein Amani, Amin Mansouri*, Nicolas Baldwin*, Martin Josifoski, Maxime Peyrard, Robert West
Preprint 2024
Multi-Domain Causal Representation Learning via Weak Distributional Invariances
Kartik Ahuja*, Amin Mansouri*, Yixin Wang
AISTATS 2024, NeurIPS 2023 Causal Representation Learning Workshop
Object-centric architectures enable efficient causal representation learning
Amin Mansouri, Jason Hartford, Yan Zhang, Yoshua Bengio
ICLR 2024, NeurIPS 2023 Causal Representation Learning Workshop
Reusable Slotwise Mechanisms
Trang Nguyen, Amin Mansouri, Kanika Madan, Nguyen Duy Khuong, Kartik Ahuja, Dianbo Liu, Yoshua Bengio
NeurIPS 2023
Object-Centric Causal Representation Learning
Amin Mansouri, Jason Hartford, Kartik Ahuja, Yoshua Bengio
NeurIPS 2022, NeurReps workshop
Amin Mansouri, Sean Spinney, Amin Memarian, Patricia Conrod, Irina Rish
Identifying Invariant and Sparse Predictors in High-dimensional Data
ICML 2021, workshop on Uncertainty & Robustness in Deep Learning, 2021
July 2025, Algorithm Discovery With LLMs: Evolutionary Search Meets Reinforcement Learning is accepted to COLM.
July 2024, Two papers accepted to ICML workshops: On Provable Length and Compositional Generalization and Symbolic Autoencoding for Self-Supervised Sequence Learning
January 2024, Two papers on Causal Representation Learning accepted at ICLR and AISTATS: Object centric architectures enable efficient causal representation learning got accepted to ICLR and Multi-Domain Causal Representation Learning via Weak Distributional Invariances got accepted to AISTATS. See you in Vienna and Valencia!
2023 October, My thesis Toward Causal Structure and Representation Learning was approved with highest distinction (Dean's Honor List).
2023 September-October, 3 papers accepted at NeurIPS main track and workshops 🎉! Reusable Slotwise Mechanisms, Object-centric architectures enable efficient causal representation learning, and Multi-Domain Causal Representation Learning via Weak Distributional Invariances in Causal Representation Learning workshop.
October 2022, Our workshop paper Object-Centric Causal Representation Learning got accepted at NeurIPS Symmetry and Geometry in Neural Representations
January 2022, I joined Mila's 2022 Entrepreneurs Cohort
I'll be joining the Data Science Lab (dlab) at EPFL (Lausanne) from May to August 2021 for the Summer@EPFL Program.
2020 September, Excited to officially start my journey at Mila, starting an M.Sc. in Computer Science
2020 September, Actually created this homepage
2020 July, Graduated from Sharif Univ. of Tech. in Electrical Engineering (Digital Systems Specialization)
2019 August, Joining Riken AIP (Fukuoka, Japan) for a summer internship
2019 February, Started playing piano
2018 July, joining ETH Zurich and Institute for Neuroinformatics for a summer internship
2017 June, Starting as the deputy director for the student association of EE Department at Sharif Univ. of Tech
2015 September, Starting my bachelor's in Electrical Engineering at Sharif Univ. of Tech.
2025 August, Finalist for Apple Scholars in AI/ML Fellowship in ML Algorithms and Architectures
2024 October, Finalist for Apple Scholars in AI/ML Fellowship in ML Theory
2024 August, My M.Sc. thesis was recognized on the Dean's Honor List
2023 January, I received University of Montreal's Scholarship for Excellence in Artificial Intelligence Research
2023 January, I was awarded EPFL IC school's prestigious Doctoral Fellowship
2021 & 2022, I was awarded the University of Montreal's scholarship for Excellence in Research
May 2021, I was awarded the Summer@EPFL fellowship.
2020 December, Microsoft scholarship for Excellence in Research.
2020 December, University of Montreal's scholarship for Responsible Development of AI
2014 September, Silver Medal in the National Olympiad on Astrophysics and Astronomy
2013 September, Bronze Medal in the National Olympiad on Astrophysics and Astronomy
