Probabilistic Contrastive Learning for Domain Adaptation

Authors: Junjie Li, Yixin Zhang, Zilei Wang, Saihui Hou, Keyu Tu, Man Zhang

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

Reproducibility Variable Result LLM Response
Research Type Experimental We conduct extensive experiments to validate the effectiveness of PCL and observe consistent performance gains on five tasks, i.e., Unsupervised/Semi-Supervised Domain Adaptation (UDA/SSDA), Semi-Supervised Learning (SSL), UDA Detection and Semantic Segmentation.
Researcher Affiliation Academia 1Beijing University of Posts and Telecommunications 2University of Science and Technology of China 3Beijing Normal University
Pseudocode No The paper does not contain any clearly labeled pseudocode or algorithm blocks.
Open Source Code Yes Code is available at https://github.com/ljjcoder/Probabilistic Contrastive-Learning.
Open Datasets Yes We evaluate our method on two standard UDA semantic segmentation tasks: GTA5 [Richter et al., 2016] Cityscapes [Cordts et al., 2016] and SYNTHIA [Ros et al., 2016] Cityscapes.
Dataset Splits No The paper refers to standard benchmarks (e.g., Domain Net, Office-Home, CIFAR-100) and specific settings like "3-shot" or "1-shot" for semi-supervised tasks, but it does not provide explicit numerical or percentage-based train/validation/test splits, nor does it cite specific predefined split methodologies.
Hardware Specification Yes Notably, the training cost of our method is much lower than CPSL-D (PCL: 1*3090, 5 days v.s. CPSL-D: 4*V100, 11 days).
Software Dependencies No The paper does not specify any software dependencies with version numbers (e.g., Python, PyTorch, TensorFlow versions).
Experiment Setup Yes For the hyperparameter in PCL, we set s = 20 in all experiments. [...] For the hyperparameter in PCL, we set s = 7 in all experiments.