CARD: Classification and Regression Diffusion Models
Authors: Xizewen Han, Huangjie Zheng, Mingyuan Zhou
NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We demonstrate the outstanding ability of CARD in conditional distribution prediction with both toy examples and real-world datasets, the experimental results on which show that CARD in general outperforms state-of-the-art methods, including Bayesian neural network-based ones that are designed for uncertainty estimation, especially when the conditional distribution of y given x is multi-modal. In addition, we utilize the stochastic nature of the generative model outputs to obtain a finer granularity in model confidence assessment at the instance level for classification tasks. Our implementation is publicly available at https://github.com/Xzw Han/CARD. |
| Researcher Affiliation | Academia | Xizewen Han Huangjie Zheng Department of Statistics and Data Sciences The University of Texas at Austin Austin, TX 78712 {xizewen.han, huangjie.zheng}@utexas.edu Mingyuan Zhou Mc Combs School of Business The University of Texas at Austin Austin, TX 78712 mingyuan.zhou@mccombs.utexas.edu |
| Pseudocode | Yes | Algorithm 1 Training (Regression) and Algorithm 2 Inference (Regression) are provided in Section 2.1. |
| Open Source Code | Yes | Our implementation is publicly available at https://github.com/Xzw Han/CARD. |
| Open Datasets | Yes | We adopt the same set of 10 UCI regression benchmark datasets (Dua and Graff, 2017) as well as the experimental protocol proposed by Hernández-Lobato and Adams (2015) and followed by Gal and Ghahramani (2016) and Lakshminarayanan et al. (2017). |
| Dataset Splits | No | The paper specifies 'multiple train-test splits with 90%/10% ratio' but does not explicitly mention or detail a separate validation dataset split. |
| Hardware Specification | Yes | The computational results reported in this paper were obtained using the Texas Advanced Computing Center (TACC) Frontera computing system... It consists of Dell PowerEdge servers, each equipped with two Intel Xeon Platinum 8280 processors and 192GB of RAM. Some experiments may have also utilized NVIDIA V100 GPUs available on the Lonestar6 system at TACC or local machines. The total computational hours consumed are approximately 4000 GPU hours for all experiments. |
| Software Dependencies | No | Our model is implemented with PyTorch (Paszke et al., 2019) and trained with Adam (Kingma and Ba, 2015; Reddi et al., 2018) optimizer. No specific version numbers for these software components are provided. |
| Experiment Setup | Yes | For the hyperparameters of CARD in both regression and classification tasks, we set the number of timesteps as T = 1000, a linear noise schedule with β1 = 10 4 and βT = 0.02, same as Ho et al. (2020). We provide a more detailed walk-through of the experimental setup, including training and network architecture, in Appendix A.8. |