1 code implementation • 19 Apr 2024 • Mostafa ElAraby, Ali Harakeh, Liam Paull
Besides the common problem of classical catastrophic forgetting in the continual learning setting, CSS suffers from the inherent ambiguity of the background, a phenomenon we refer to as the "background shift'', since pixels labeled as background could correspond to future classes (forward background shift) or previous classes (backward background shift).
no code implementations • 3 Apr 2024 • Manfred Diaz, Liam Paull, Andrea Tacchetti
Teacher-Student Curriculum Learning (TSCL) is a curriculum learning framework that draws inspiration from human cultural transmission and learning.
no code implementations • 29 Mar 2024 • Luke Rowe, Roger Girgis, Anthony Gosselin, Bruno Carrez, Florian Golemo, Felix Heide, Liam Paull, Christopher Pal
With this dataset, we train a return-conditioned multi-agent behaviour model that allows for fine-grained manipulation of agent behaviours by modifying the desired returns for the various reward components.
no code implementations • 22 Dec 2023 • Mostafa ElAraby, Sabyasachi Sahoo, Yann Pequignot, Paul Novello, Liam Paull
To build this space, GROOD relies on class prototypes together with a prototype that specifically captures OOD characteristics.
no code implementations • 23 Oct 2023 • Zhen Liu, Yao Feng, Yuliang Xiu, Weiyang Liu, Liam Paull, Michael J. Black, Bernhard Schölkopf
Recent work has focused on the former, and methods for reconstructing open surfaces do not support fast reconstruction with material and lighting or unconditional generative modelling.
no code implementations • 28 Sep 2023 • Qiao Gu, Alihusein Kuwajerwala, Sacha Morin, Krishna Murthy Jatavallabhula, Bipasha Sen, Aditya Agarwal, Corban Rivera, William Paul, Kirsty Ellis, Rama Chellappa, Chuang Gan, Celso Miguel de Melo, Joshua B. Tenenbaum, Antonio Torralba, Florian Shkurti, Liam Paull
We demonstrate the utility of this representation through a number of downstream planning tasks that are specified through abstract (language) prompts and require complex reasoning over spatial and semantic concepts.
1 code implementation • 14 Mar 2023 • Zhen Liu, Yao Feng, Michael J. Black, Derek Nowrouzezahrai, Liam Paull, Weiyang Liu
We consider the task of generating realistic 3D shapes, which is useful for a variety of applications such as automatic scene generation and physical simulation.
no code implementations • 7 Mar 2023 • Sacha Morin, Miguel Saavedra-Ruiz, Liam Paull
A fundamental task in robotics is to navigate between two locations.
1 code implementation • 14 Feb 2023 • Krishna Murthy Jatavallabhula, Alihusein Kuwajerwala, Qiao Gu, Mohd Omama, Tao Chen, Alaa Maalouf, Shuang Li, Ganesh Iyer, Soroush Saryazdi, Nikhil Keetha, Ayush Tewari, Joshua B. Tenenbaum, Celso Miguel de Melo, Madhava Krishna, Liam Paull, Florian Shkurti, Antonio Torralba
ConceptFusion leverages the open-set capabilities of today's foundation models pre-trained on internet-scale data to reason about concepts across modalities such as natural language, images, and audio.
no code implementations • CVPR 2023 • Anas Mahmoud, Jordan S. K. Hu, Tianshu Kuai, Ali Harakeh, Liam Paull, Steven L. Waslander
However, image-to point representation learning for autonomous driving datasets faces two main challenges: 1) the abundance of self-similarity, which results in the contrastive losses pushing away semantically similar point and image regions and thus disturbing the local semantic structure of the learned representations, and 2) severe class imbalance as pretraining gets dominated by over-represented classes.
no code implementations • 6 Jan 2023 • Vincent Mai, Philippe Maisonneuve, Tianyu Zhang, Hadi Nekoei, Liam Paull, Antoine Lesage-Landry
To integrate high amounts of renewable energy resources, electrical power grids must be able to cope with high amplitude, fast timescale variations in power generation.
Multi-agent Reinforcement Learning reinforcement-learning +1
no code implementations • 24 Nov 2022 • Ali Harakeh, Jordan Hu, Naiqing Guan, Steven L. Waslander, Liam Paull
A common approach to model uncertainty is to choose a parametric distribution and fit the data to it using maximum likelihood estimation.
no code implementations • 14 Nov 2022 • Amir Rasouli, Randy Goebel, Matthew E. Taylor, Iuliia Kotseruba, Soheil Alizadeh, Tianpei Yang, Montgomery Alban, Florian Shkurti, Yuzheng Zhuang, Adam Scibior, Kasra Rezaee, Animesh Garg, David Meger, Jun Luo, Liam Paull, Weinan Zhang, Xinyu Wang, Xi Chen
The proposed competition supports methodologically diverse solutions, such as reinforcement learning (RL) and offline learning methods, trained on a combination of naturalistic AD data and open-source simulation platform SMARTS.
no code implementations • 11 Oct 2022 • Ruixiang Zhang, Tong Che, Boris Ivanovic, Renhao Wang, Marco Pavone, Yoshua Bengio, Liam Paull
Humans are remarkably good at understanding and reasoning about complex visual scenes.
no code implementations • 20 Jul 2022 • Weiyang Liu, Zhen Liu, Liam Paull, Adrian Weller, Bernhard Schölkopf
This paper considers the problem of unsupervised 3D object reconstruction from in-the-wild single-view images.
1 code implementation • 11 Jul 2022 • Christopher Agia, Krishna Murthy Jatavallabhula, Mohamed Khodeir, Ondrej Miksik, Vibhav Vineet, Mustafa Mukadam, Liam Paull, Florian Shkurti
3D scene graphs (3DSGs) are an emerging description; unifying symbolic, topological, and metric scene representations.
no code implementations • 7 Mar 2022 • Miguel Saavedra-Ruiz, Sacha Morin, Liam Paull
In this work, we consider the problem of learning a perception model for monocular robot navigation using few annotated images.
1 code implementation • ICLR 2022 • Vincent Mai, Kaustubh Mani, Liam Paull
In model-free deep reinforcement learning (RL) algorithms, using noisy value estimates to supervise policy evaluation and optimization is detrimental to the sample efficiency.
no code implementations • NeurIPS 2021 • Weiyang Liu, Zhen Liu, Hanchen Wang, Liam Paull, Bernhard Schölkopf, Adrian Weller
In this paper, we consider the problem of iterative machine teaching, where a teacher provides examples sequentially based on the current iterative learner.
no code implementations • 16 Oct 2021 • Rey Reza Wiyatno, Anqi Xu, Liam Paull
Commonly, learning-based topological navigation approaches produce a local policy while preserving some loose connectivity of the space through a topological map.
1 code implementation • 29 Sep 2021 • Anthony Courchesne, Andrea Censi, Liam Paull
We propose the relative predictive PU to assess the predictive ability of a proxy domain and the learning PU to quantify the usefulness of a proxy as a tool to generate learning data.
no code implementations • 28 Sep 2021 • Dhaivat Bhatt, Kaustubh Mani, Dishank Bansal, Krishna Murthy, Hanju Lee, Liam Paull
While modern deep neural networks are performant perception modules, performance (accuracy) alone is insufficient, particularly for safety-critical robotic applications such as self-driving vehicles.
no code implementations • 9 Jul 2021 • Vincent Mai, Waleed Khamies, Liam Paull
In many situations however, the labelling process is able to estimate the variance of such distribution for each label, which can be used as an additional information to mitigate this impact.
no code implementations • ICLR 2021 • Krishna Murthy Jatavallabhula, Miles Macklin, Florian Golemo, Vikram Voleti, Linda Petrini, Martin Weiss, Breandan Considine, Jerome Parent-Levesque, Kevin Xie, Kenny Erleben, Liam Paull, Florian Shkurti, Derek Nowrouzezahrai, Sanja Fidler
We consider the problem of estimating an object's physical properties such as mass, friction, and elasticity directly from video sequences.
no code implementations • ICLR Workshop SSL-RL 2021 • Manfred Diaz, Liam Paull, Pablo Samuel Castro
We offer a novel approach to balance exploration and exploitation in reinforcement learning (RL).
no code implementations • 1 Jan 2021 • Vincent Mai, Waleed Khamies, Liam Paull
In this work, we consider this setting and additionally assume that the label generating process is able to provide us with a quantity for the role of each label in the misalignment between the datasets, which we consider to be privileged information.
2 code implementations • NeurIPS 2020 • Gunshi Gupta, Karmesh Yadav, Liam Paull
The continual learning problem involves training models with limited capacity to perform well on a set of an unknown number of sequentially arriving tasks.
no code implementations • 9 Sep 2020 • Jacopo Tani, Andrea F. Daniele, Gianmarco Bernasconi, Amaury Camus, Aleksandar Petrov, Anthony Courchesne, Bhairav Mehta, Rohit Suri, Tomasz Zaluska, Matthew R. Walter, Emilio Frazzoli, Liam Paull, Andrea Censi
As robotics matures and increases in complexity, it is more necessary than ever that robot autonomy research be reproducible.
3 code implementations • ICML Workshop LifelongML 2020 • Gunshi Gupta, Karmesh Yadav, Liam Paull
The continual learning problem involves training models with limited capacity to perform well on a set of an unknown number of sequentially arriving tasks.
1 code implementation • CVPR 2021 • Weiyang Liu, Rongmei Lin, Zhen Liu, James M. Rehg, Liam Paull, Li Xiong, Le Song, Adrian Weller
The inductive bias of a neural network is largely determined by the architecture and the training algorithm.
3 code implementations • NeurIPS 2020 • Tong Che, Ruixiang Zhang, Jascha Sohl-Dickstein, Hugo Larochelle, Liam Paull, Yuan Cao, Yoshua Bengio
To make that practical, we show that sampling from this modified density can be achieved by sampling in latent space according to an energy-based model induced by the sum of the latent prior log-density and the discriminator output score.
no code implementations • 19 Feb 2020 • Bhairav Mehta, Tristan Deleu, Sharath Chandra Raparthy, Chris J. Pal, Liam Paull
However, specifically in the case of meta-reinforcement learning (meta-RL), we can show that gradient-based meta-learners are sensitive to task distributions.
1 code implementation • 18 Feb 2020 • Sharath Chandra Raparthy, Bhairav Mehta, Florian Golemo, Liam Paull
Goal-directed Reinforcement Learning (RL) traditionally considers an agent interacting with an environment, prescribing a real-valued reward to an agent proportional to the completion of some goal.
1 code implementation • 23 Oct 2019 • Krishna Murthy Jatavallabhula, Soroush Saryazdi, Ganesh Iyer, Liam Paull
Blending representation learning approaches with simultaneous localization and mapping (SLAM) systems is an open question, because of their highly modular and complex nature.
2 code implementations • ICML 2020 • Zijun Zhang, Ruixiang Zhang, Zongpeng Li, Yoshua Bengio, Liam Paull
We therefore propose to map both the generated and target distributions to a latent space using the encoder of a standard autoencoder, and train the generator (or decoder) to match the target distribution in the latent space.
2 code implementations • 9 Apr 2019 • Bhairav Mehta, Manfred Diaz, Florian Golemo, Christopher J. Pal, Liam Paull
Our experiments show that domain randomization may lead to suboptimal, high-variance policies, which we attribute to the uniform sampling of environment parameters.
1 code implementation • 5 Mar 2019 • Sai Krishna, Keehong Seo, Dhaivat Bhatt, Vincent Mai, Krishna Murthy, Liam Paull
Traditional approaches to this use an information-theoretic criterion for action selection and hand-crafted perceptual models.
no code implementations • 11 Oct 2018 • Homanga Bharadhwaj, Zihan Wang, Yoshua Bengio, Liam Paull
Learning effective visuomotor policies for robots purely from data is challenging, but also appealing since a learning-based system should not require manual tuning or calibration.
no code implementations • 11 Apr 2018 • Ganesh Iyer, J. Krishna Murthy, Gunshi Gupta, K. Madhava Krishna, Liam Paull
We show that using a noisy teacher, which could be a standard VO pipeline, and by designing a loss term that enforces geometric consistency of the trajectory, we can train accurate deep models for VO that do not require ground-truth labels.
2 code implementations • 19 Apr 2017 • Beipeng Mu, Shih-Yuan Liu, Liam Paull, John Leonard, Jonathan How
The \textit{data association} and \textit{simultaneous localization and mapping} (SLAM) problems are, individually, well-studied in the literature.