no code implementations • 26 Mar 2024 • Adrien Lafage, Mathieu Barbier, Gianni Franchi, David Filliat
Accurate trajectory forecasting is crucial for the performance of various systems, such as advanced driver-assistance systems and self-driving vehicles.
no code implementations • 9 Oct 2023 • David Brellmann, Eloïse Berthier, David Filliat, Goran Frehse
We identify the ratio between the number of parameters and the number of visited states as a crucial factor and define over-parameterization as the regime when it is larger than one.
1 code implementation • 27 Sep 2023 • Gianni Franchi, Marwane Hariat, Xuanlong Yu, Nacim Belkhir, Antoine Manzanera, David Filliat
Current deep neural networks (DNNs) for autonomous driving computer vision are typically trained on specific datasets that only involve a single type of data and urban scenes.
no code implementations • 14 Jun 2023 • Tom Dupuis, Jaonary Rabarisoa, Quoc-Cuong Pham, David Filliat
In this work, we tackle the problem of robust visual control at its core and present VIBR (View-Invariant Bellman Residuals), a method that combines multi-view training and invariant prediction to reduce out-of-distribution (OOD) generalization gap for RL based visuomotor control.
1 code implementation • 20 Jul 2022 • Gianni Franchi, Xuanlong Yu, Andrei Bursuc, Emanuel Aldea, Severine Dubuisson, David Filliat
Predictive uncertainty estimation is essential for deploying Deep Neural Networks in real-world autonomous systems.
3 code implementations • 2 Mar 2022 • Gianni Franchi, Xuanlong Yu, Andrei Bursuc, Angel Tena, Rémi Kazmierczak, Séverine Dubuisson, Emanuel Aldea, David Filliat
However, disentangling the different types and sources of uncertainty is non trivial for most datasets, especially since there is no ground truth for uncertainty.
1 code implementation • 17 Feb 2022 • Rémi Kazmierczak, Gianni Franchi, Nacim Belkhir, Antoine Manzanera, David Filliat
Several metrics exist to quantify the similarity between images, but they are inefficient when it comes to measure the similarity of highly distorted images.
no code implementations • 29 Sep 2021 • David Brellmann, Goran Frehse, David Filliat
To the best of our knowledge, this is the first reported application of Fourier features in Deep RL.
1 code implementation • 17 Sep 2021 • Zhaorun Chen, Siqi Fan, Yuan Tan, Liang Gong, Binhao Chen, Te Sun, David Filliat, Natalia Díaz-Rodríguez, Chengliang Liu
Firstly, We engage RL loss to assist in updating SRL model so that the states can evolve to meet the demand of RL and maintain a good physical interpretation.
no code implementations • 5 Jul 2021 • Hugo Caselles-Dupré, Michael Garcia-Ortiz, David Filliat
We introduce SCOD (Sensory Commutativity Object Detection), an active method for movable and immovable object detection.
no code implementations • 5 Jul 2021 • Hugo Caselles-Dupré, Michael Garcia-Ortiz, David Filliat
We conclude that the problem of supervised affordance segmentation is included in the problem of object segmentation and argue that better benchmarks for affordance learning should include action capacities.
1 code implementation • 20 May 2021 • Florence Carton, David Filliat, Jaonary Rabarisoa, Quoc Cuong Pham
In recent years, we have witnessed increasingly high performance in the field of autonomous end-to-end driving.
2 code implementations • 24 Apr 2021 • Natalia Díaz-Rodríguez, Alberto Lamas, Jules Sanchez, Gianni Franchi, Ivan Donadello, Siham Tabik, David Filliat, Policarpo Cruz, Rosana Montes, Francisco Herrera
We tackle such problem by considering the symbolic knowledge is expressed in form of a domain expert knowledge graph.
no code implementations • 2 Apr 2021 • Thomas Rojat, Raphaël Puget, David Filliat, Javier Del Ser, Rodolphe Gelin, Natalia Díaz-Rodríguez
Most of state of the art methods applied on time series consist of deep learning methods that are too complex to be interpreted.
no code implementations • 13 May 2020 • Stephane Doncieux, Nicolas Bredeche, Léni Le Goff, Benoît Girard, Alexandre Coninx, Olivier Sigaud, Mehdi Khamassi, Natalia Díaz-Rodríguez, David Filliat, Timothy Hospedales, A. Eiben, Richard Duro
Robots are still limited to controlled conditions, that the robot designer knows with enough details to endow the robot with the appropriate models or behaviors.
no code implementations • 13 Feb 2020 • Hugo Caselles-Dupré, Michael Garcia-Ortiz, David Filliat
In such case, for autonomous embodied agents with first-person sensors, perception can be learned end-to-end to solve particular tasks.
no code implementations • 6 Dec 2019 • Timothée Lesort, Andrei Stoian, David Filliat
In most machine learning algorithms, training data is assumed to be independent and identically distributed (iid).
no code implementations • 11 Jul 2019 • René Traoré, Hugo Caselles-Dupré, Timothée Lesort, Te Sun, Guanghang Cai, Natalia Díaz-Rodríguez, David Filliat
In multi-task reinforcement learning there are two main challenges: at training time, the ability to learn different policies with a single model; at test time, inferring which of those policies applying without an external signal.
no code implementations • 29 Jun 2019 • Timothée Lesort, Vincenzo Lomonaco, Andrei Stoian, Davide Maltoni, David Filliat, Natalia Díaz-Rodríguez
An important challenge for machine learning is not necessarily finding solutions that work in the real world but rather finding stable algorithms that can learn in real world.
no code implementations • 11 Jun 2019 • René Traoré, Hugo Caselles-Dupré, Timothée Lesort, Te Sun, Natalia Díaz-Rodríguez, David Filliat
We focus on the problem of teaching a robot to solve tasks presented sequentially, i. e., in a continual learning scenario.
1 code implementation • NeurIPS 2019 • Hugo Caselles-Dupré, Michael Garcia-Ortiz, David Filliat
Finding a generally accepted formal definition of a disentangled representation in the context of an agent behaving in an environment is an important challenge towards the construction of data-efficient autonomous agents.
no code implementations • 25 Feb 2019 • Hugo Caselles-Dupré, Michael Garcia-Ortiz, David Filliat
As the environment changes, the aim is to efficiently compress the sensory state's information without losing past knowledge, and then use Reinforcement Learning on the resulting features for efficient policy learning.
5 code implementations • 24 Jan 2019 • Antonin Raffin, Ashley Hill, René Traoré, Timothée Lesort, Natalia Díaz-Rodríguez, David Filliat
Scaling end-to-end reinforcement learning to control real robots from vision presents a series of challenges, in particular in terms of sample efficiency.
1 code implementation • ICLR 2019 • Timothée Lesort, Hugo Caselles-Dupré, Michael Garcia-Ortiz, Andrei Stoian, David Filliat
We experiment with sequential tasks on three commonly used benchmarks for Continual Learning (MNIST, Fashion MNIST and CIFAR10).
no code implementations • 31 Oct 2018 • Natalia Díaz-Rodríguez, Vincenzo Lomonaco, David Filliat, Davide Maltoni
Continual learning consists of algorithms that learn from a stream of data/tasks continuously and adaptively thought time, enabling the incremental development of ever more complex knowledge and skills.
no code implementations • 29 Oct 2018 • Timothée Lesort, Alexander Gepperth, Andrei Stoian, David Filliat
We present a new replay-based method of continual classification learning that we term "conditional replay" which generates samples and labels together by sampling from a distribution conditioned on the class.
no code implementations • 9 Oct 2018 • Hugo Caselles-Dupré, Michael Garcia-Ortiz, David Filliat
As the environment changes, the aim is to efficiently compress the sensory state's information without losing past knowledge.
5 code implementations • 25 Sep 2018 • Antonin Raffin, Ashley Hill, René Traoré, Timothée Lesort, Natalia Díaz-Rodríguez, David Filliat
State representation learning aims at learning compact representations from raw observations in robotics and control applications.
no code implementations • 12 Sep 2018 • Clément Pinard, Laure Chevalley, Antoine Manzanera, David Filliat
We propose a depth map inference system from monocular videos based on a novel dataset for navigation that mimics aerial footage from gimbal stabilized monocular camera in rigid scenes.
no code implementations • 12 Sep 2018 • Clément Pinard, Laure Chevalley, Antoine Manzanera, David Filliat
Using a neural network architecture for depth map inference from monocular stabilized videos with application to UAV videos in rigid scenes, we propose a multi-range architecture for unconstrained UAV flight, leveraging flight data from sensors to make accurate depth maps for uncluttered outdoor environment.
no code implementations • 12 Sep 2018 • Clément Pinard, Laure Chevalley, Antoine Manzanera, David Filliat
We then present results on a synthetic dataset that we believe to be more representative of typical UAV scenes.
no code implementations • 3 Sep 2018 • Hugo Caselles-Dupré, Louis Annabi, Oksana Hagen, Michael Garcia-Ortiz, David Filliat
Flatland is a simple, lightweight environment for fast prototyping and testing of reinforcement learning agents.
no code implementations • 28 Jun 2018 • Timothée Lesort, Andrei Stoain, Jean-François Goudou, David Filliat
By comparing results with different generated datasets we are able to classify and compare generative models.
no code implementations • 2 Apr 2018 • Celine Craye, Timothee Lesort, David Filliat, Jean-Francois Goudou
On the other hand, we investigate an autonomous exploration technique to efficiently learn such a saliency model.
1 code implementation • 12 Feb 2018 • Timothée Lesort, Natalia Díaz-Rodríguez, Jean-François Goudou, David Filliat
State representation learning (SRL) focuses on a particular kind of representation learning where learned features are in low dimension, evolve through time, and are influenced by actions of an agent.
no code implementations • ICLR 2018 • Timothée Lesort, Florian Bordes, Jean-Francois Goudou, David Filliat
This mixture of real and generated data is thus used to train a classifier which is afterwards tested on a given labeled test dataset.
no code implementations • 15 Sep 2017 • Timothée Lesort, Mathieu Seurin, Xinrui Li, Natalia Díaz Rodríguez, David Filliat
We reproduce this simplification process using a neural network to build a low dimensional state representation of the world from images acquired by a robot.
no code implementations • 10 Dec 2015 • Olivier Sigaud, Clément Masson, David Filliat, Freek Stulp
Gated networks are networks that contain gating connections, in which the outputs of at least two neurons are multiplied.