arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2406.06976cs.LGcs.AI
收藏

用于结构化表示学习的基于离散字典的分解层

Discrete Dictionary-based Decomposition Layer for Structured Representation Learning

Kyungpook National University(庆北国立大学);ALI Co., Ltd.(ALI有限公司)

Taewon Park, Hyun-Chul Kim, Minho Lee

AI总结:

针对基于张量积表示(TPR)的神经网络难以分解未见数据的问题,提出基于离散可学习键值字典的D3即插即用层,利用训练先验生成结构化TPR表示,在更少额外参数下提升多种TPR模型的系统泛化能力,合成任务表现优于基线。

摘要:

神经符号神经网络已被广泛研究,旨在将符号运算与神经网络相结合,从而提升系统泛化能力。具体而言,Tensor Product Representation(TPR,张量积表示)框架通过在向量空间中编码数据的符号结构,使神经网络能够执行可微的符号运算。然而,基于TPR的神经网络通常难以将未见过的数据分解为结构化的TPR表示,这削弱了其符号运算能力。为解决这一分解问题,我们提出了Discrete Dictionary-based Decomposition(D3,基于离散字典的分解)层,旨在增强基于TPR的模型的分解能力。D3采用离散的、可学习的键值字典,这些字典经过训练以捕捉分解运算所需的关键符号特征。它利用训练过程中习得的先验知识,通过将输入数据映射到这些字典中预先学习到的符号特征,生成结构化的TPR表示。D3是一种简单的即插即用层,无需修改即可无缝集成到任何基于TPR的模型中。我们的实验结果表明,D3显著提升了多种基于TPR的模型的系统泛化能力,同时所需的额外参数更少。值得注意的是,在需要对未见过的组合数据进行系统分解的合成任务上,D3的表现优于基线模型。

英文摘要:

Neuro-symbolic neural networks have been extensively studied to integrate symbolic operations with neural networks, thereby improving systematic generalization. Specifically, Tensor Product Representation (TPR) framework enables neural networks to perform differentiable symbolic operations by encoding the symbolic structure of data within vector spaces. However, TPR-based neural networks often struggle to decompose unseen data into structured TPR representations, undermining their symbolic operations. To address this decomposition problem, we propose a Discrete Dictionary-based Decomposition (D3) layer designed to enhance the decomposition capabilities of TPR-based models. D3 employs discrete, learnable key-value dictionaries trained to capture symbolic features essential for decomposition operations. It leverages the prior knowledge acquired during training to generate structured TPR representations by mapping input data to pre-learned symbolic features within these dictionaries. D3 is a straightforward drop-in layer that can be seamlessly integrated into any TPR-based model without modifications. Our experimental results demonstrate that D3 significantly improves the systematic generalization of various TPR-based models while requiring fewer additional parameters. Notably, D3 outperforms baseline models on the synthetic task that demands the systematic decomposition of unseen combinatorial data.

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。

↑