update

_pages/cv.md

@@ -1,6 +1,6 @@
 ---
 layout: archive
-title: "CV"
+title: " "
 permalink: /cv/
 author_profile: true
 redirect_from:

_publications/2022-03-01-reactive stepping.md

@@ -6,7 +6,7 @@ video: 'https://www.youtube.com/watch?v=HLx6CHfpmBM'
 permalink: /publication/2022-03-01-reactive-stepping
 excerpt: '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.'
 date: 2022-03-01
-venue: '2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)'
+venue: 'IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)'
 paperurl: 'https://arxiv.org/abs/2203.01148'
 citation: 'Duburcq et al. (2022). &quot;Reactive Stepping for Humanoid Robots using Reinforcement Learning: Application to Standing Push Recovery on the Exoskeleton Atalante.&quot; <i>IROS22</i>.'
 ---

_publications/2023-06-19-qp layer.md

@@ -5,7 +5,7 @@ authors: 'Antoine Bambade, <b>Fabian Schramm</b>, Adrien Taylor, Justin Carpenti
 permalink: /publication/2023-06-19-qp-layer
 excerpt: 'This paper presents primal-dual augmented Lagrangian techniques for computing derivatives of both feasible and infeasible QPs.'
 date: 2023-06-19
-venue: 'Twelfth International Conference on Learning Representations'
+venue: 'Twelfth International Conference on Learning Representations (ICLR)'
 paperurl: 'https://hal.laas.fr/PRAIRIE-IA/hal-04133055v1'
 citation: 'Bambade et al. (2023). &quot;Leveraging augmented-Lagrangian techniques for differentiating over infeasible quadratic programs in machine learning.&quot; <i>ICLR24</i>.'
 ---

_publications/2024-01-22-rs.md

@@ -0,0 +1,14 @@
+---
+title: "Leveraging Randomized Smoothing for Optimal Control of Nonsmooth Dynamical Systems"
+collection: publications
+authors: 'Quentin Le Lidec, <b>Fabian Schramm</b>, Louis Montaut, Cordelia Schmid, Ivan Laptev, Justin Carpentier'
+permalink: /publication/2024-01-22-rs
+excerpt: '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.'
+date: 2024-05-01
+venue: 'Nonlinear Analysis: Hybrid Systems, International Federation of Automatic Control (IFAC) journal'
+paperurl: 'https://arxiv.org/abs/2203.03986'
+citation: 'Le Lidec et al. (2024). &quot;Leveraging Randomized Smoothing for Optimal Control of Nonsmooth Dynamical Systems.&quot; <i>NAHS24</i>.'
+---
+Optimal control (OC) algorithms such as Differential Dynamic Programming (DDP) take advantage of the derivatives of the dynamics to efficiently control physical systems. Yet, in the presence of nonsmooth dynamical systems, such class of algorithms are likely to fail due, for instance, to the presence of discontinuities in the dynamics derivatives or because of non-informative gradient. On the contrary, reinforcement learning (RL) algorithms have shown better empirical results in scenarios exhibiting non-smooth effects (contacts, frictions, etc). Our approach leverages recent works on randomized smoothing (RS) to tackle non-smoothness issues commonly encountered in optimal control, and provides key insights on the interplay between RL and OC through the prism of RS methods. This naturally leads us to introduce the randomized Differential Dynamic Programming (R-DDP) algorithm accounting for deterministic but non-smooth dynamics in a very sample-efficient way. The experiments demonstrate that our method is able to solve classic robotic problems with dry friction and frictional contacts, where classical OC algorithms are likely to fail and RL algorithms require in practice a prohibitive number of samples to find an optimal solution.
+
+[Download paper here](https://arxiv.org/pdf/2203.03986.pdf)