Understanding in Artificial Intelligence
arXiv:2101.06573
Abstract
Current Artificial Intelligence (AI) methods, most based on deep learning, have facilitated progress in several fields, including computer vision and natural language understanding. The progress of these AI methods is measured using benchmarks designed to solve challenging tasks, such as visual question answering. A question remains of how much understanding is leveraged by these methods and how appropriate are the current benchmarks to measure understanding capabilities. To answer these questions, we have analysed existing benchmarks and their understanding capabilities, defined by a set of understanding capabilities, and current research streams. We show how progress has been made in benchmark development to measure understanding capabilities of AI methods and we review as well how current methods develop understanding capabilities.
28 pages, 282 references
References in corpus (25)
- Deep Learning in Neural Networks: An Overview
- Cooperative Game Theory Approaches for Network Partitioning
- Language Models are Few-Shot Learners
- Microsoft COCO Captions: Data Collection and Evaluation Server
- DRAW: A Recurrent Neural Network For Image Generation
- Seq2SQL: Generating Structured Queries from Natural Language using Reinforcement Learning
- The Next Decade in AI: Four Steps Towards Robust Artificial Intelligence
- A Multi-World Approach to Question Answering about Real-World Scenes based on Uncertain Input
- MONet: Unsupervised Scene Decomposition and Representation
- Logic Tensor Networks: Deep Learning and Logical Reasoning from Data and Knowledge
- Neural-Symbolic Computing: An Effective Methodology for Principled Integration of Machine Learning and Reasoning
- Logical Neural Networks
- TerpreT: A Probabilistic Programming Language for Program Induction
- NewsQA: A Machine Comprehension Dataset
- A Wrapped Normal Distribution on Hyperbolic Space for Gradient-Based Learning
- Image-based table recognition: data, model, and evaluation
- Differentiable Programs with Neural Libraries
- Hyperbolic Disk Embeddings for Directed Acyclic Graphs
- Improving Question Answering over Incomplete KBs with Knowledge-Aware Reader
- Deep Learning for Learning Graph Representations
- An Explicit Local and Global Representation Disentanglement Framework with Applications in Deep Clustering and Unsupervised Object Detection
- Cracking the Contextual Commonsense Code: Understanding Commonsense Reasoning Aptitude of Deep Contextual Representations
- Can a Gorilla Ride a Camel? Learning Semantic Plausibility from Text
- Ranking and Selecting Multi-Hop Knowledge Paths to Better Predict Human Needs
- Winograd Schema - Knowledge Extraction Using Narrative Chains