Fabian Schramm

Fabian Schramm

Ph.D. Student at Inria and ENS Paris

Publications

Survival Reinforcement Learning: Toward Scalable Self-Supervised RL

Franki Nguimatsia-Tiofack, Fabian Schramm, Théotime Le Hellard, Justin Carpentier

Published in arXiv preprint, 2026

We introduce Survival Reinforcement Learning (SRL), an online classification-based alternative to contrastive RL that extends the survival value learning framework by maximizing the agent’s dwell time at target goals, matching state-of-the-art CRL on manipulation and outperforming it by 2x to 8x on long-horizon locomotion tasks.

SVL: Goal-Conditioned Reinforcement Learning as Survival Learning

Franki Nguimatsia Tiofack, Fabian Schramm, Théotime Le Hellard, Justin Carpentier

Published in International Conference on Machine Learning (ICML), 2026

We present SVL (Survival Value Learning), a probabilistic approach that reframes goal-conditioned reinforcement learning as a survival learning problem by modeling the time-to-goal as a probability distribution.

Variance-Reduced Model Predictive Path Integral via Quadratic Model Approximation

Fabian Schramm, Franki Nguimatsia Tiofack, Nicolas Perrin-Gilbert, Marc Toussaint, Justin Carpentier

Published in Robotics: Science and Systems (RSS), 2026

Sampling-based controllers, such as Model Predictive Path Integral (MPPI) methods, offer substantial flexibility but often suffer from high variance and low sample efficiency. To address these challenges, we introduce a hybrid variance-reduced MPPI framework that integrates a prior model into the sampling process.

Guided Flow Policy: Learning from High-Value Actions in Offline Reinforcement Learning

Franki Nguimatsia Tiofack*, Théotime Le Hellard*, Fabian Schramm*, Nicolas Perrin-Gilbert, Justin Carpentier
* Equal contribution

Published in Fourteenth International Conference on Learning Representations (ICLR), 2026

We introduce Guided Flow Policy (GFP), which couples a multi-step flow-matching policy with a distilled one-step actor to focus on learning from high-value actions in offline reinforcement learning.

First-order Sobolev Reinforcement Learning

Fabian Schramm, Nicolas Perrin-Gilbert, Justin Carpentier

Published in Differentiable Systems and Scientific Machine Learning, EurIPS 2025, 2025

We propose a refinement of temporal-difference learning that enforces first-order Bellman consistency by training the value function to match both Bellman targets and their derivatives.

Reference-Free Sampling-Based Model Predictive Control

Fabian Schramm, Pierre Fabre, Nicolas Perrin-Gilbert, Justin Carpentier

Published in IEEE International Conference on Robotics and Automation (ICRA), 2025

We present a sampling-based model predictive control framework that enables emergent locomotion without relying on handcrafted gait patterns or predefined contact sequences.

Leveraging Randomized Smoothing for Optimal Control of Nonsmooth Dynamical Systems

Quentin Le Lidec, Fabian Schramm, Louis Montaut, Cordelia Schmid, Ivan Laptev, Justin Carpentier

Published in Nonlinear Analysis: Hybrid Systems, International Federation of Automatic Control (IFAC) journal, 2024

This paper presents randomized smoothing to tackle non-smoothness issues commonly encountered in optimal control and provides key insights on the interplay between Reinforcement Learning and Optimal Control.

Reactive Stepping for Humanoid Robots using Reinforcement Learning: Application to Standing Push Recovery on the Exoskeleton Atalante

Alexis Duburcq, Fabian Schramm, Guilhem Boéris, Nicolas Bredeche, Yann Chevaleyre

Published in IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2022

Video

This paper presents a reinforcement learning framework capable of learning robust standing push recovery for bipedal robots with a smooth out-of-the-box transfer to reality, requiring only instantaneous proprioceptive observations.