Efficient Codes for Inverse Dynamics During Walking
Authors: Leif Johnson, Dana Ballard
AAAI 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In this paper, we explore the use of efficient codes for representing information relevant to human movements during locomotion. Specifically, we apply motion capture data to a physical model of the human skeleton to compute joint angles (inverse kinematics) and joint torques (inverse dynamics); then, by treating the resulting paired dataset as a supervised regression problem, we investigate the effect of sparsity in mapping from angles to torques. The results of our investigation suggest that sparse codes can indeed represent salient features of both the kinematic and dynamic views of human locomotion movements. |
| Researcher Affiliation | Academia | Leif Johnson and Dana H. Ballard Computer Science Department The University of Texas at Austin {leif,dana}@cs.utexas.edu |
| Pseudocode | No | The paper does not include any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper mentions using and linking to third-party open-source packages like 'theanets' and 'scikit-learn', but it does not state that the authors' own source code for the methodology described in the paper is available or provide a link to it. |
| Open Datasets | No | The paper describes the creation of a custom dataset from motion capture data ("To gather movement data, we used a 16 camera Phasespace 1 motion-capture system...") but does not provide any concrete access information (link, DOI, repository, or citation to a public version) for this dataset. |
| Dataset Splits | No | The paper states, "Finally, we set aside 10% of the whitened data to use for testing", implying the rest is for training, but it does not explicitly specify a separate validation split or its percentage for hyperparameter tuning or early stopping, although early stopping is mentioned. |
| Hardware Specification | No | The paper does not provide any specific details about the hardware (e.g., CPU, GPU models, memory, cloud instance types) used to run the experiments. |
| Software Dependencies | No | The paper mentions the "theanets Python package 3" (with '3' being a footnote number to a URL, not a version number like 0.x.x) and "scikit-learn (Pedregosa et al. 2011)" with a citation but no version number. This does not meet the requirement of providing specific version numbers for key software components. |
| Experiment Setup | No | The paper mentions general aspects like "stochastic gradient descent and early stopping" for training and that "the size of the codebook W and the degree of sparsity λ can be varied", but it does not provide concrete hyperparameter values (e.g., learning rate, batch size) or specific values of W and λ used for the reported results. |