no code implementations • 6 Jan 2024 • Rafael Rafailov, Kyle Hatch, Victor Kolev, John D. Martin, Mariano Phielipp, Chelsea Finn
We study the problem of offline pre-training and online fine-tuning for reinforcement learning from high-dimensional observations in the context of realistic robot tasks.
no code implementations • 4 Oct 2023 • Raj Ghugare, Santiago Miret, Adriana Hugessen, Mariano Phielipp, Glen Berseth
Reinforcement learning (RL) over text representations can be effective for finding high-value policies that can search over graphs.
1 code implementation • 23 Nov 2022 • Austin Cheng, Andy Cai, Santiago Miret, Gustavo Malkomes, Mariano Phielipp, Alán Aspuru-Guzik
We introduce Group SELFIES, a molecular string representation that leverages group tokens to represent functional groups or entire substructures while maintaining chemical robustness guarantees.
no code implementations • 17 Jun 2022 • Brandon Trabucco, Mariano Phielipp, Glen Berseth
Ours is the first reinforcement learning algorithm that can train a policy to generalize to new agent morphologies without requiring a description of the agent's morphology in advance.
2 code implementations • 22 May 2022 • Anurag Koul, Mariano Phielipp, Alan Fern
Decision makers often wish to use offline historical data to compare sequential-action policies at various world states.
1 code implementation • 29 Mar 2022 • Kourosh Hakhamaneshi, Marcel Nassar, Mariano Phielipp, Pieter Abbeel, Vladimir Stojanović
We show that pretraining GNNs on prediction of output node voltages can encourage learning representations that can be adapted to new unseen topologies or prediction of new circuit level properties with up to 10x more sample efficiency compared to a randomly initialized model.
1 code implementation • 31 Jan 2022 • Hassam Sheikh, Kizza Frisbee, Mariano Phielipp
Application of ensemble of neural networks is becoming an imminent tool for advancing the state-of-the-art in deep reinforcement learning algorithms.
no code implementations • ICLR 2022 • Hassam Sheikh, Mariano Phielipp, Ladislau Boloni
In this paper, we describe Maximize Ensemble Diversity in Reinforcement Learning (MED-RL), a set of regularization methods inspired from the economics and consensus optimization to improve diversity in the ensemble-based deep reinforcement learning methods by encouraging inequality between the networks during training.
no code implementations • 15 Jun 2021 • Varun Kumar Vijay, Hassam Sheikh, Somdeb Majumdar, Mariano Phielipp
With these techniques, we show that we can reduce communication by 75% with no loss of performance.
no code implementations • 14 Jun 2021 • Santiago Miret, Vui Seng Chua, Mattias Marder, Mariano Phielipp, Nilesh Jain, Somdeb Majumdar
In this work, we present a flexible and scalable framework for automated mixed-precision quantization that concurrently optimizes task performance, memory compression, and compute savings through multi-objective evolutionary computing.
no code implementations • 1 Jan 2021 • Martin A Bertran, Guillermo Sapiro, Mariano Phielipp
Deep Reinforcement Learning (DRL) can distill behavioural policies from sensory input that solve complex tasks, however, the policies tend to be task-specific and sample inefficient, requiring a large number of interactions with the environment that may be costly or impractical for many real world applications.
no code implementations • NeurIPS 2020 • Martin Bertran, Natalia Martinez, Mariano Phielipp, Guillermo Sapiro
Agents trained via deep reinforcement learning (RL) routinely fail to generalize to unseen environments, even when these share the same underlying dynamics as the training levels.
1 code implementation • 2 Nov 2020 • Martin Bertran, Natalia Martinez, Mariano Phielipp, Guillermo Sapiro
Agents trained via deep reinforcement learning (RL) routinely fail to generalize to unseen environments, even when these share the same underlying dynamics as the training levels.
1 code implementation • NeurIPS 2020 • Simon Stepputtis, Joseph Campbell, Mariano Phielipp, Stefan Lee, Chitta Baral, Heni Ben Amor
Imitation learning is a popular approach for teaching motor skills to robots.
no code implementations • 31 May 2020 • Ajay Kumar Tanwani, Pierre Sermanet, Andy Yan, Raghav Anand, Mariano Phielipp, Ken Goldberg
We demonstrate the use of this representation to imitate surgical suturing motions from publicly available videos of the JIGSAWS dataset.
no code implementations • 5 Dec 2019 • Siyu Zhou, Mariano Phielipp, Jorge A. Sefair, Sara I. Walker, Heni Ben Amor
In this paper, we propose SwarmNet -- a neural network architecture that can learn to predict and imitate the behavior of an observed swarm of agents in a centralized manner.
no code implementations • 26 Nov 2019 • Simon Stepputtis, Joseph Campbell, Mariano Phielipp, Chitta Baral, Heni Ben Amor
In this work we propose a novel end-to-end imitation learning approach which combines natural language, vision, and motion information to produce an abstract representation of a task, which in turn is used to synthesize specific motion controllers at run-time.
no code implementations • 25 Sep 2019 • Simon Stepputtis, Joseph Campbell, Mariano Phielipp, Chitta Baral, Heni Ben Amor
In this work we propose a novel end-to-end imitation learning approach which combines natural language, vision, and motion information to produce an abstract representation of a task, which in turn can be used to synthesize specific motion controllers at run-time.
no code implementations • 29 Jun 2019 • Zach Dwiel, Madhavun Candadai, Mariano Phielipp
The use of robotics in controlled environments has flourished over the last several decades and training robots to perform tasks using control strategies developed from dynamical models of their hardware have proven very effective.
1 code implementation • NeurIPS 2019 • Yiming Ding, Carlos Florensa, Mariano Phielipp, Pieter Abbeel
Designing rewards for Reinforcement Learning (RL) is challenging because it needs to convey the desired task, be efficient to optimize, and be easy to compute.
no code implementations • 4 May 2019 • Zach Dwiel, Madhavun Candadai, Mariano Phielipp, Arjun K. Bansal
Hierarchy in reinforcement learning agents allows for control at multiple time scales yielding improved sample efficiency, the ability to deal with long time horizons and transferability of sub-policies to tasks outside the training distribution.