Search Results for author: Nicholas Roy

Found 24 papers, 3 papers with code

PRompt Optimization in Multi-Step Tasks (PROMST): Integrating Human Feedback and Preference Alignment

1 code implementation13 Feb 2024 Yongchao Chen, Jacob Arkin, Yilun Hao, Yang Zhang, Nicholas Roy, Chuchu Fan

However, realistic tasks for agents are multi-step and introduce new challenges: (1) Prompt content is likely to be more extensive and complex, making it more difficult for LLMs to analyze errors, (2) the impact of an individual step is difficult to evaluate, and (3) different people may have varied preferences about task execution.

Language Modelling Large Language Model

AutoTAMP: Autoregressive Task and Motion Planning with LLMs as Translators and Checkers

3 code implementations10 Jun 2023 Yongchao Chen, Jacob Arkin, Charles Dawson, Yang Zhang, Nicholas Roy, Chuchu Fan

Rather than using LLMs to directly plan task sub-goals, we instead perform few-shot translation from natural language task descriptions to an intermediate task representation that can then be consumed by a TAMP algorithm to jointly solve the task and motion plan.

Motion Planning Task and Motion Planning +1

Structured Latent Variable Models for Articulated Object Interaction

no code implementations26 May 2023 Emily Liu, Michael Noseworthy, Nicholas Roy

In this paper, we investigate a scenario in which a robot learns a low-dimensional representation of a door given a video of the door opening or closing.

Object

Active Learning of Abstract Plan Feasibility

no code implementations1 Jul 2021 Michael Noseworthy, Caris Moses, Isaiah Brand, Sebastian Castro, Leslie Kaelbling, Tomás Lozano-Pérez, Nicholas Roy

Long horizon sequential manipulation tasks are effectively addressed hierarchically: at a high level of abstraction the planner searches over abstract action sequences, and when a plan is found, lower level motion plans are generated.

Active Learning

Online Descriptor Enhancement via Self-Labelling Triplets for Visual Data Association

no code implementations6 Nov 2020 Yorai Shaoul, Katherine Liu, Kyel Ok, Nicholas Roy

We show that self-labelling challenging triplets--choosing positive examples separated by large temporal distances and negative examples close in the descriptor space--improves the quality of the learned descriptors for the multi-object tracking task.

Image Classification Multi-Object Tracking +2

Leveraging Past References for Robust Language Grounding

no code implementations CONLL 2019 Subhro Roy, Michael Noseworthy, Rohan Paul, Daehyung Park, Nicholas Roy

We therefore reframe the grounding problem from the perspective of coreference detection and propose a neural network that detects when two expressions are referring to the same object.

Object Referring Expression +1

Admissible Abstractions for Near-optimal Task and Motion Planning

no code implementations3 Jun 2018 William Vega-Brown, Nicholas Roy

We define an admissibility condition for abstractions expressed using angelic semantics and show that these conditions allow us to accelerate planning while preserving the ability to find the optimal motion plan.

Motion Planning Task and Motion Planning

FLaME: Fast Lightweight Mesh Estimation Using Variational Smoothing on Delaunay Graphs

no code implementations ICCV 2017 W. Nicholas Greene, Nicholas Roy

We propose a lightweight method for dense online monocular depth estimation capable of reconstructing 3D meshes on computationally constrained platforms.

Monocular Depth Estimation

PROBE-GK: Predictive Robust Estimation using Generalized Kernels

no code implementations1 Aug 2017 Valentin Peretroukhin, William Vega-Brown, Nicholas Roy, Jonathan Kelly

Many algorithms in computer vision and robotics make strong assumptions about uncertainty, and rely on the validity of these assumptions to produce accurate and consistent state estimates.

Bayesian Inference

Structural Return Maximization for Reinforcement Learning

no code implementations12 May 2014 Joshua Joseph, Javier Velez, Nicholas Roy

Batch Reinforcement Learning (RL) algorithms attempt to choose a policy from a designer-provided class of policies given a fixed set of training data.

Learning Theory reinforcement-learning +1

Modelling Observation Correlations for Active Exploration and Robust Object Detection

no code implementations18 Jan 2014 Javier Velez, Garrett Hemann, Albert S. Huang, Ingmar Posner, Nicholas Roy

In particular, the performance of detection algorithms is commonly sensitive to the position of the sensor relative to the objects in the scene.

object-detection Robust Object Detection +1

Efficient Planning under Uncertainty with Macro-actions

no code implementations16 Jan 2014 Ruijie He, Emma Brunskill, Nicholas Roy

We also demonstrate our algorithm being used to control a real robotic helicopter in a target monitoring experiment, which suggests that our approach has practical potential for planning in real-world, large partially observable domains where a multi-step lookahead is required to achieve good performance.

Batch-iFDD for Representation Expansion in Large MDPs

no code implementations26 Sep 2013 Alborz Geramifard, Thomas J. Walsh, Nicholas Roy, Jonathan How

Matching pursuit (MP) methods are a promising class of feature construction algorithms for value function approximation.

Cannot find the paper you are looking for? You can Submit a new open access paper.