Abstraction and Analogy-Making in Artificial Intelligence
arXiv:2102.10717 · doi:10.1111/nyas.14619
Abstract
Conceptual abstraction and analogy-making are key abilities underlying humans' abilities to learn, reason, and robustly adapt their knowledge to new domains. Despite of a long history of research on constructing AI systems with these abilities, no current AI system is anywhere close to a capability of forming humanlike abstractions or analogies. This paper reviews the advantages and limitations of several approaches toward this goal, including symbolic methods, deep learning, and probabilistic program induction. The paper concludes with several proposals for designing challenge tasks and evaluation measures in order to make quantifiable and generalizable progress in this area.
arXiv commentRevised version. 30 pages, 9 figures. To appear in Annals of the New York Academy of Sciences
References in corpus (23)
- Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
- Prototypical Networks for Few-shot Learning
- Shortcut Learning in Deep Neural Networks
- GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding
- The Neuro-Symbolic Concept Learner: Interpreting Scenes, Words, and Sentences From Natural Supervision
- Measuring abstract reasoning in neural networks
- Synthesizing Programs for Images using Reinforced Adversarial Learning
- Five Points to Check when Comparing Visual Perception in Humans and Machines
- Learning to Make Analogies by Contrasting Abstract Relational Structure
- Improving Generalization for Abstract Reasoning Tasks Using Disentangled Feature Representations
- Learning Perceptual Inference by Contrasting
- Grounded Language Learning Fast and Slow
- Abstract Reasoning with Distracting Features
- The Scattering Compositional Learner: Discovering Objects, Attributes, Relationships in Analogical Reasoning
- Bongard-LOGO: A New Benchmark for Human-Level Concept Learning and Reasoning
- VISALOGY: Answering Visual Analogy Questions
- Solving Raven's Progressive Matrices with Neural Networks
- Solving Bongard Problems with a Visual Language and Pragmatic Reasoning
- Transforming task representations to perform novel tasks
- Emergent Symbols through Binding in External Memory
- Attention on Abstract Visual Reasoning
- Generating new concepts with hybrid neuro-symbolic models
- Abstract Spatial-Temporal Reasoning via Probabilistic Abduction and Execution
Cited by in corpus (6)
- A Review of Emerging Research Directions in Abstract Visual Reasoning
- Deep Learning Methods for Abstract Visual Reasoning: A Survey on Raven's Progressive Matrices
- The Duality of Data and Knowledge Across the Three Waves of AI
- A Neural Approach for Detecting Morphological Analogies
- Abstraction, Reasoning and Deep Learning: A Study of the "Look and Say" Sequence
- Toward Building Science Discovery Machines