Conformal Prediction for Uncertainty-Aware Planning with Diffusion Dynamics Model

Authors: Jiankai Sun, Yiqi Jiang, Jianing Qiu, Parth Nobel, Mykel J Kochenderfer, Mac Schwager

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

Reproducibility Variable Result LLM Response
Research Type Experimental We evaluate our algorithm on various planning tasks and model-based offline reinforcement learning tasks and show that it reduces the uncertainty of the learned trajectory prediction model.
Researcher Affiliation Academia Jiankai Sun Stanford University jksun@stanford.edu Yiqi Jiang Stanford University yqjiang@stanford.edu Jianing Qiu Imperial College London jianing.qiu17@imperial.ac.uk Parth Talpur Nobel Stanford University ptnobel@stanford.edu Mykel Kochenderfer Stanford University mykel@stanford.edu Mac Schwager Stanford University schwager@stanford.edu
Pseudocode Yes Algorithm 1: Plan CP: Conformal Prediction for Planning with Diffusion Dynamics Models
Open Source Code No The paper does not provide any explicit statement or link regarding the availability of its source code.
Open Datasets Yes Finally, we evaluate the ability to recover an effective single-task controller from heterogeneous data of varying quality using the D4RL Benchmark [75].
Dataset Splits Yes We have split the dataset D into three parts Dtrain, Dcal, and Dtest for training, calibration, and testing, respectively.
Hardware Specification No The paper does not provide specific hardware details (such as CPU, GPU models, or memory) used for running its experiments.
Software Dependencies No The paper mentions using the Adam optimizer but does not provide specific version numbers for any software dependencies.
Experiment Setup Yes We set the uncertainty weight to λuncertainty = 5 and the failure probability to α = 0.1. To optimize the model, we use the Adam [73, 74] optimizer with a learning rate of 2 10 4. We train the diffusion dynamics model on the training set Dtrain for 2 105 iterations.