首页/论文/2007.04212

arXiv 2007.042122020-07-08cs.LGcs.AIcs.LOstat.ML

The Scattering Compositional Learner: Discovering Objects, Attributes, Relationships in Analogical Reasoning

University of Toronto（多伦多大学）；Vector Institute（Vector研究所）

Yuhuai Wu, Honghua Dong, Roger Grosse, Jimmy Ba

英文摘要：

In this work, we focus on an analogical reasoning task that contains rich compositional structures, Raven's Progressive Matrices (RPM). To discover compositional structures of the data, we propose the Scattering Compositional Learner (SCL), an architecture that composes neural networks in a sequence. Our SCL achieves state-of-the-art performance on two RPM datasets, with a 48.7% relative improvement on Balanced-RAVEN and 26.4% on PGM over the previous state-of-the-art. We additionally show that our model discovers compositional representations of objects' attributes (e.g., shape color, size), and their relationships (e.g., progression, union). We also find that the compositional representation makes the SCL significantly more robust to test-time domain shifts and greatly improves zero-shot generalization to previously unseen analogies.

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。

阅读原文

阅读 PDF ↗

arXiv 页面 ↗HTML 原文 ↗

本文内容

英文摘要相关论文

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。

继续探索

相关论文

LacMaterial：大型语言模型作为材料发现的类比化学家

LacMaterial: Large Language Models as Analogical Chemists for Materials Discovery

National Research Council Canada（加拿大国家研究理事会）

本研究提出LacMaterial框架，利用大型语言模型通过跨领域类比检索和领域内类比模板生成新型电池材料，超越传统掺杂替代，产出更优候选，展现类比驱动的科学创新潜力。

2025-10-25cs.AIcs.LG

Discovering the Compositional Structure of Vector Representations with Role Learning Networks

Johns Hopkins University（约翰斯·霍普金斯大学）；Microsoft Research（微软研究院）；New York University（纽约大学）

How can neural networks perform so well on compositional tasks even though they lack explicit compositional representations? We use a novel analysis technique called ROLE to show that recurrent neural networks perform well on such tasks by converging to solutions which implicitly represent symbolic structure. This method uncovers a symbolic structure which, when properly embedded in vector space, closely approximates the encodings of a standard seq2seq network trained to perform the compositional SCAN task. We verify the causal importance of the discovered symbolic structure by showing that, w…

2020-11-16cs.CLcs.LGstat.ML

DeepGAR: Deep Graph Learning for Analogical Reasoning

Emory University（埃默里大学）；George Mason University（乔治梅森大学）；Tianjin University（天津大学）；NEC Labs America（NEC美国实验室）

Analogical reasoning is the process of discovering and mapping correspondences from a target subject to a base subject. As the most well-known computational method of analogical reasoning, Structure-Mapping Theory (SMT) abstracts both target and base subjects into relational graphs and forms the cognitive process of analogical reasoning by finding a corresponding subgraph (i.e., correspondence) in the target graph that is aligned with the base graph. However, incorporating deep learning for SMT is still under-explored due to several obstacles: 1) the combinatorial complexity of searching for t…

2022-11-19cs.AIcs.LGcs.SI

发现能够组合泛化的模块化解决方案

Discovering modular solutions that generalize compositionally

ETH Zurich（苏黎世联邦理工学院）；IDEAS NCBR；Imperial College London（帝国理工学院）；Google Research（谷歌研究院）；Google DeepMind（谷歌DeepMind）

研究师生设置下模块化系统发现隐藏组合结构的条件，从理论上证明超网络可仅从演示识别模块，并通过实验验证元学习能发现组合泛化的模块化策略。

2024-03-25cs.LGcs.NE

Transformer 见，Transformer 作：复制作为学习类比推理的中间步骤

Transformer See, Transformer Do: Copying as an Intermediate Step in Learning Analogical Reasoning

University of Amsterdam（阿姆斯特丹大学）

本文研究了通过Meta-Learning for Compositionality训练的Transformer在类比推理任务中的表现，发现引入复制任务可提升模型对字母字符串类比的学习能力，并通过异质数据集训练提高了对新字母集的泛化能力。

2026-08-21cs.CLcs.LG

Visual analogy: Deep learning versus compositional models

University of California, Los Angeles（加利福尼亚大学洛杉矶分校）；Johns Hopkins University（约翰斯·霍普金斯大学）

Is analogical reasoning a task that must be learned to solve from scratch by applying deep learning models to massive numbers of reasoning problems? Or are analogies solved by computing similarities between structured representations of analogs? We address this question by comparing human performance on visual analogies created using images of familiar three-dimensional objects (cars and their subregions) with the performance of alternative computational models. Human reasoners achieved above-chance accuracy for all problem types, but made more errors in several conditions (e.g., when relevant…

2021-05-14cs.AIcs.LG

更多相关论文收起更多论文

Solving Visual Analogies Using Neural Algorithmic Reasoning

BITS Pilani, K.K. Birla Goa Campus（印度比尔拉理工学院果阿校区）；TCS Research, New Delhi（塔塔咨询服务公司研究院）

We consider a class of visual analogical reasoning problems that involve discovering the sequence of transformations by which pairs of input/output images are related, so as to analogously transform future inputs. This program synthesis task can be easily solved via symbolic search. Using a variation of the `neural analogical reasoning' approach of (Velickovic and Blundell 2021), we instead search for a sequence of elementary neural network transformations that manipulate distributed representations derived from a symbolic space, to which input images are directly encoded. We evaluate the exte…

2021-11-19cs.AIcs.LG

ARC：利用组合表示进行VRPs上的跨问题学习

ARC: Leveraging Compositional Representations for Cross-Problem Learning on VRPs

LG AI Research（LG人工智能研究院）

ARC通过组合学习方法，解耦属性表示以提升VRPs跨问题学习的泛化能力，实现多场景下的高效适应与性能优化。

2025-12-21cs.AIcs.LG

自适应提示结构分解：一种自我发现和优化组合提示程序的框架

Adaptive Prompt Structure Factorization: A Framework for Self-Discovering and Optimizing Compositional Prompt Programs

The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳））；Taobao and Tmall Group（淘宝天猫集团）

本文提出aPSF框架，通过语义因素发现和单因素更新优化提示程序，提升大语言模型的推理可靠性，实验显示在多个基准上准确率提升2.16个百分点，优化成本降低45-87%。

2026-04-08cs.CLcs.LG

Discovering Human-Object Interaction Concepts via Self-Compositional Learning

The University of Sydney（悉尼大学）；JD Explore Academy（京东探索研究院）

A comprehensive understanding of human-object interaction (HOI) requires detecting not only a small portion of predefined HOI concepts (or categories) but also other reasonable HOI concepts, while current approaches usually fail to explore a huge portion of unknown HOI concepts (i.e., unknown but reasonable combinations of verbs and objects). In this paper, 1) we introduce a novel and challenging task for a comprehensive HOI understanding, which is termed as HOI Concept Discovery; and 2) we devise a self-compositional learning framework (or SCL) for HOI concept discovery. Specifically, we main…

2022-07-24cs.CVcs.LGcs.RO

Why PairDiff works? -- A Mathematical Analysis of Bilinear Relational Compositional Operators for Analogy Detection

University of Liverpool（利物浦大学）；Japan Advanced Institute of Science and Technology（日本先端科学技术大学院大学）

Representing the semantic relations that exist between two given words (or entities) is an important first step in a wide-range of NLP applications such as analogical reasoning, knowledge base completion and relational information retrieval. A simple, yet surprisingly accurate method for representing a relation between two words is to compute the vector offset (\PairDiff) between their corresponding word embeddings. Despite the empirical success, it remains unclear as to whether \PairDiff is the best operator for obtaining a relational representation from word embeddings. We conduct a theoreti…

2017-12-15cs.AIcs.CLcs.LG

Compositional Structure Learning for Sequential Video Data

Seoul National University（首尔国立大学）；Kakao Brain

Conventional sequential learning methods such as Recurrent Neural Networks (RNNs) focus on interactions between consecutive inputs, i.e. first-order Markovian dependency. However, most of sequential data, as seen with videos, have complex temporal dependencies that imply variable-length semantic flows and their compositions, and those are hard to be captured by conventional methods. Here, we propose Temporal Dependency Networks (TDNs) for learning video data by discovering these complex structures of the videos. The TDNs represent video as a graph whose nodes and edges correspond to frames of …

2019-07-03cs.CVcs.LGeess.IV

Cut-Based Graph Learning Networks to Discover Compositional Structure of Sequential Video Data

Seoul National University（首尔国立大学）；Kakao Brain

Conventional sequential learning methods such as Recurrent Neural Networks (RNNs) focus on interactions between consecutive inputs, i.e. first-order Markovian dependency. However, most of sequential data, as seen with videos, have complex dependency structures that imply variable-length semantic flows and their compositions, and those are hard to be captured by conventional methods. Here, we propose Cut-Based Graph Learning Networks (CB-GLNs) for learning video data by discovering these complex structures of the video. The CB-GLNs represent video data as a graph, with nodes and edges correspon…

2020-01-17cs.CVcs.LGeess.IV

基于注意力的张量积表示迭代分解

Attention-based Iterative Decomposition for Tensor Product Representation

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

针对张量积表示（TPR）在系统泛化任务中分解不完整的问题，提出基于注意力的迭代分解（AID）模块，通过竞争性注意力机制增强结构表示的分解，显著提升了基于TPR模型的泛化性能与表示质量。

2024-06-03cs.AIcs.LG