Active representation learning for general task space with applications in robotics

Authors: Yifang Chen, Yingbing Huang, Simon S. Du, Kevin G. Jamieson, Guanya Shi

NeurIPS 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Finally, we demonstrate different instantiations of our meta algorithm in synthetic datasets and robotics problems, from pendulum simulations to real-world drone flight datasets. On average, our algorithms outperform baselines by 20% 70%.
Researcher Affiliation Academia 1 Paul G. Allen School of Computer Science & Engineering University of Washington, Seattle,WA {yifangc, ssdu, jamieson, guanyas}@cs.washington.edu 2 University of Illinois Urbana-Champaign, Champaign, IL {yh21}@illinois.edu 3 Robotics Institute, Carnegie Mellon University, Pittsburgh, PA {{guanyas }@andrew.cmu.edu
Pseudocode Yes Algorithm 1 Active multi-task representation learning (general templates)
Open Source Code Yes 1Code in https://github.com/cloudways X/ALMulti Task_Robotics
Open Datasets Yes Real-world drone flight dataset [7]. The Neural-Fly dataset [7] includes real flight trajectories using two different drones in various wind conditions.
Dataset Splits No The paper does not provide specific details on training, validation, or test dataset splits (e.g., percentages or exact counts) for any of the datasets used.
Hardware Specification No No specific hardware details (e.g., GPU models, CPU types, or memory specifications) used for running the experiments were provided in the paper.
Software Dependencies No The paper mentions using a 'gradient-descent joint training oracle' but does not specify any particular software packages or their version numbers required for reproducibility.
Experiment Setup No The paper mentions using a 'gradient-descent joint training oracle' but does not provide specific hyperparameter values (e.g., learning rate, batch size, number of epochs) or other detailed training configurations in the main text.