no code implementations • 18 Sep 2024 • Stefano Ferraro, Pietro Mazzaglia, Tim Verbelen, Bart Dhoedt, Sai Rajeswar
As we analyze the causes of this limitation, we identify the cause of underperformance in the way current world models represent crucial positional information, especially about the target's goal specification for object positioning tasks.
1 code implementation • 26 Jun 2024 • Pietro Mazzaglia, Tim Verbelen, Bart Dhoedt, Aaron Courville, Sai Rajeswar
In this work, we overcome these problems by presenting multimodal-foundation world models, able to connect and align the representation of foundation VLMs with the latent space of generative world models for RL, without any language annotations.
1 code implementation • 6 Jun 2024 • Pietro Mazzaglia, Nicholas Backshall, Xiao Ma, Stephen James
Joint space and task space control are the two dominant action modes for controlling robot arms within the robot learning literature.
no code implementations • 6 May 2024 • Pietro Mazzaglia, Taco Cohen, Daniel Dijkman
Robotic affordances, providing information about what actions can be taken in a given situation, can aid robotic manipulation.
no code implementations • 5 Jul 2023 • Stefano Ferraro, Pietro Mazzaglia, Tim Verbelen, Bart Dhoedt
Understanding the world in terms of objects and the possible interplays with them is an important cognition ability, especially in robotics manipulation, where many tasks require robot-object interactions.
no code implementations • 4 May 2023 • Mattijs Baert, Pietro Mazzaglia, Sam Leroux, Pieter Simoens
To address this challenge, we propose a novel method that utilizes the principle of maximum causal entropy to learn constraints and an optimal policy that adheres to these constraints, using demonstrations of agents that abide by the constraints.
no code implementations • 7 Feb 2023 • Toon Van de Maele, Tim Verbelen, Pietro Mazzaglia, Stefano Ferraro, Bart Dhoedt
Representing a scene and its constituent objects from raw sensory data is a core ability for enabling robots to interact with their environment.
1 code implementation • 23 Nov 2022 • Pietro Mazzaglia, Tim Verbelen, Bart Dhoedt, Alexandre Lacoste, Sai Rajeswar
Unsupervised skill learning aims to learn a rich repertoire of behaviors without external supervision, providing artificial agents with the ability to control and influence the environment.
1 code implementation • 24 Sep 2022 • Sai Rajeswar, Pietro Mazzaglia, Tim Verbelen, Alexandre Piché, Bart Dhoedt, Aaron Courville, Alexandre Lacoste
In this work, we study the URLB and propose a new method to solve it, using unsupervised model-based RL, for pre-training the agent, and a task-aware fine-tuning strategy combined with a new proposed hybrid planner, Dyna-MPC, to adapt the agent for downstream tasks.
no code implementations • 16 Sep 2022 • Stefano Ferraro, Toon Van de Maele, Pietro Mazzaglia, Tim Verbelen, Bart Dhoedt
Recently, deep learning methods have been proposed to learn a hidden state space structure purely from data, alleviating the experimenter from this tedious design task, but resulting in an entangled, non-interpreteable state space.
no code implementations • 19 Aug 2022 • Daria de Tinguy, Pietro Mazzaglia, Tim Verbelen, Bart Dhoedt
When studying unconstrained behaviour and allowing mice to leave their cage to navigate a complex labyrinth, the mice exhibit foraging behaviour in the labyrinth searching for rewards, returning to their home cage now and then, e. g. to drink.
no code implementations • 13 Jul 2022 • Pietro Mazzaglia, Tim Verbelen, Ozan Çatal, Bart Dhoedt
The free energy principle, and its corollary active inference, constitute a bio-inspired theory that assumes biological agents act to remain in a restricted set of preferred states of the world, i. e., they minimize their free energy.
1 code implementation • NeurIPS 2021 • Pietro Mazzaglia, Tim Verbelen, Bart Dhoedt
In this work, we propose a contrastive objective for active inference that strongly reduces the computational burden in learning the agent's generative model and planning future actions.
1 code implementation • 22 Apr 2021 • Samuel T. Wauthier, Pietro Mazzaglia, Ozan Çatal, Cedric De Boom, Tim Verbelen, Bart Dhoedt
Historically, artificial intelligence has drawn much inspiration from neuroscience to fuel advances in the field.
2 code implementations • ICLR Workshop SSL-RL 2021 • Pietro Mazzaglia, Ozan Catal, Tim Verbelen, Bart Dhoedt
The human intrinsic desire to pursue knowledge, also known as curiosity, is considered essential in the process of skill acquisition.