arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2406.19121cs.LGcs.AIcs.SC
收藏

利用VSA分布式表示学习溯因推理

Towards Learning Abductive Reasoning using VSA Distributed Representations

IBM Research – Zurich(IBM研究院苏黎世分部);ETH Zürich(苏黎世联邦理工学院)

Giacomo Camposampiero, Michael Hersche, Aleksandar Terzić, Roger Wattenhofer, Abu Sebastian, Abbas Rahimi

AI总结:

本文提出ARLC模型,通过新颖的溯因推理训练目标在I-RAVEN上实现最先进准确率,兼具可解释性、抗遗忘和迁移学习能力。

摘要:

我们提出了具有上下文感知能力的溯因规则学习器(ARLC),这是一个基于Learn-VRF解决抽象推理任务的模型。ARLC采用了一种新颖且适用范围更广的溯因推理训练目标,从而在解决瑞文渐进矩阵(RPM)时具有更好的可解释性和更高的准确率。ARLC既支持编程领域知识,也支持学习数据分布背后的规则。我们在I-RAVEN数据集上对ARLC进行了评估,展示了其在分布内和分布外(未见过的属性-规则对)测试中的最先进准确率。尽管参数量少了几个数量级,ARLC仍超越了神经符号和连接主义基线,包括大型语言模型。我们展示了ARLC对编程后训练的鲁棒性,通过在已编程知识的基础上逐步从示例中学习,这只会提高其性能,而不会导致对已编程解决方案的灾难性遗忘。我们验证了ARLC从2x2 RPM星座到未见星座的无缝迁移学习。我们的代码可在https://github.com/IBM/abductive-rule-learner-with-context-awareness获取。

英文摘要:

We introduce the Abductive Rule Learner with Context-awareness (ARLC), a model that solves abstract reasoning tasks based on Learn-VRF. ARLC features a novel and more broadly applicable training objective for abductive reasoning, resulting in better interpretability and higher accuracy when solving Raven's progressive matrices (RPM). ARLC allows both programming domain knowledge and learning the rules underlying a data distribution. We evaluate ARLC on the I-RAVEN dataset, showcasing state-of-the-art accuracy across both in-distribution and out-of-distribution (unseen attribute-rule pairs) tests. ARLC surpasses neuro-symbolic and connectionist baselines, including large language models, despite having orders of magnitude fewer parameters. We show ARLC's robustness to post-programming training by incrementally learning from examples on top of programmed knowledge, which only improves its performance and does not result in catastrophic forgetting of the programmed solution. We validate ARLC's seamless transfer learning from a 2x2 RPM constellation to unseen constellations. Our code is available at https://github.com/IBM/abductive-rule-learner-with-context-awareness.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑