Fourier Transporter: Bi-Equivariant Robotic Manipulation in 3D

Authors: Haojie Huang, Owen Lewis Howell, Dian Wang, Xupeng Zhu, Robert Platt, Robin Walters

ICLR 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Tests on the RLbench benchmark achieve state-of-the-art results across various tasks.
Researcher Affiliation Academia Northeastern University, Boston, MA 02115, USA
Pseudocode Yes Algorithm 1 Four Tran inference
Open Source Code No The paper provides a 'Project website' link, but it does not explicitly state that the source code for the described methodology is available there, nor does it link directly to a code repository.
Open Datasets Yes Tests on the RLbench benchmark achieve state-of-the-art results across various tasks. (James et al. (2020))
Dataset Splits No The paper mentions evaluating 'the best evaluation across the training process' but does not specify a distinct validation dataset split (e.g., by percentage or count) for reproducibility.
Hardware Specification Yes Tests were performed on NVIDIA 3090 GPU.
Software Dependencies No The paper mentions software components like UNet and Adam optimizer, but does not provide specific version numbers for these or other critical software dependencies required for reproduction.
Experiment Setup Yes We train our method with {1, 5, 10} demonstrations and train the baselines with 10 demonstrations on each task individually. All methods are trained for 15K SGD steps, and we evaluate them on 25 unseen configurations every 5K steps. [...] The model is trained using the Adam optimizer with fixed learning rate=1e 4.