Safe Abductive Learning in the Presence of Inaccurate Rules

Authors: Xiao-Wen Yang, Jie-Jing Shao, Wei-Wei Tu, Yu-Feng Li, Wang-Zhou Dai, Zhi-Hua Zhou

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

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments on diverse tasks show that our method can tolerate at least twice as many inaccurate rules as accurate ones and achieve highly competitive performance while other methods can t.
Researcher Affiliation Collaboration 1National Key Laboratory for Novel Software Technology, Nanjing University, China School of Artificial Intelligence, Nanjing University, China 24Paradigm Inc., Beijing, China {yangxw, shaojj, liyf, daiwz, zhouzh}@lamda.nju.edu.cn, tuww.cn@gmail.com
Pseudocode No The paper does not contain any structured pseudocode or algorithm blocks.
Open Source Code No The paper does not provide any specific links or explicit statements about releasing source code for the described methodology.
Open Datasets Yes MNIST Addition This task was first introduced by Deep Problog (Manhaeve et al. 2018) which contains two subtasks, Single-digit and Multi-digit. The input of the first subtask is a pair of MNIST images (Le Cun et al. 1998), and the output is the sum of the individual digits.
Dataset Splits No The paper does not explicitly provide training/test/validation dataset splits with percentages, sample counts, or references to predefined splits.
Hardware Specification Yes All experiments are repeated five times with Nvidia Tesla V100 GPU.
Software Dependencies No The paper mentions using a pre-trained BERT model and Le Net-5 as perception models, but it does not specify software dependencies with version numbers (e.g., Python, PyTorch, TensorFlow versions).
Experiment Setup Yes The σ is the hyperparameter that selects with high confidence and we set σ = 0.99 in all our experiments.