KANDINSKYPatterns -- An experimental exploration environment for Pattern Analysis and Machine Intelligence
arXiv:2103.00519
Abstract
Machine intelligence is very successful at standard recognition tasks when having high-quality training data. There is still a significant gap between machine-level pattern recognition and human-level concept learning. Humans can learn under uncertainty from only a few examples and generalize these concepts to solve new problems. The growing interest in explainable machine intelligence, requires experimental environments and diagnostic tests to analyze weaknesses in existing approaches to drive progress in the field. In this paper, we discuss existing diagnostic tests and test data sets such as CLEVR, CLEVERER, CLOSURE, CURI, Bongard-LOGO, V-PROM, and present our own experimental environment: The KANDINSKYPatterns, named after the Russian artist Wassily Kandinksy, who made theoretical contributions to compositivity, i.e. that all perceptions consist of geometrically elementary individual components. This was experimentally proven by Hubel &Wiesel in the 1960s and became the basis for machine learning approaches such as the Neocognitron and the even later Deep Learning. While KANDINSKYPatterns have computationally controllable properties on the one hand, bringing ground truth, they are also easily distinguishable by human observers, i.e., controlled patterns can be described by both humans and algorithms, making them another important contribution to international research in machine intelligence.
arXiv comment12 pages, submitted to IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), currently under review
References in corpus (26)
- Microsoft COCO Captions: Data Collection and Evaluation Server
- VQA: Visual Question Answering
- Neural-Symbolic VQA: Disentangling Reasoning from Vision and Language Understanding
- Higher-Order Explanations of Graph Neural Networks via Relevant Walks
- MONet: Unsupervised Scene Decomposition and Representation
- Measuring abstract reasoning in neural networks
- Compositional Attention Networks for Machine Reasoning
- Learning to Reason: End-to-End Module Networks for Visual Question Answering
- Analysing Mathematical Reasoning Abilities of Neural Models
- Generating Visual Explanations
- Contrastive Learning of Structured World Models
- CLEVRER: CoLlision Events for Video REpresentation and Reasoning
- Visually Grounded Neural Syntax Acquisition
- C-VQA: A Compositional Split of the Visual Question Answering (VQA) v1.0 Dataset
- ShapeWorld - A new test methodology for multimodal language understanding
- CLOSURE: Assessing Systematic Generalization of CLEVR Models
- RAVEN: A Dataset for Relational and Analogical Visual rEasoNing
- Kandinsky Patterns
- Systematic Generalization: What Is Required and Can It Be Learned?
- Bongard-LOGO: A New Benchmark for Human-Level Concept Learning and Reasoning
- Towards causal generative scene models via competition of experts
- Object-Centric Diagnosis of Visual Reasoning
- CURI: A Benchmark for Productive Concept Learning Under Uncertainty
- Visually Grounded Compound PCFGs
- Contextual Grounding of Natural Language Entities in Images
- What is Learned in Visually Grounded Neural Syntax Acquisition