Projection: A Mechanism for Human-like Reasoning in Artificial Intelligence
Not on the map. Why?
Abstract (arXiv)
Artificial Intelligence systems cannot yet match human abilities to apply knowledge to situations that vary from what they have been programmed for, or trained for. In visual object recognition methods of inference exploiting top-down information (from a model) have been shown to be effective for recognising entities in difficult conditions. Here this type of inference, called `projection', is shown to be a key mechanism to solve the problem of applying knowledge to varied or challenging situations, across a range of AI domains, such as vision, robotics, or language. Finally the relevance of projection to tackling the commonsense knowledge problem is discussed.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 24 of its 103 references that are in the map. First posted on arXiv 24 Mar 2021.
Builds on
24 of 103 references.
References · 24
- 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
- 2018ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustnesscited by 1,430
- 2017Bottom-Up and Top-Down Attention for Image Captioning and Visual Question Answeringcited by 1,310
- 2021Perceiver: General Perception with Iterative Attentioncited by 1,089
- 2017AVA: A Video Dataset of Spatio-temporally Localized Atomic Visual Actionscited by 449
- 2017Pose Guided Person Image Generationcited by 251
- 2021How to represent part-whole hierarchies in a neural networkcited by 123
- 2017CAN: Creative Adversarial Networks, Generating "Art" by Learning About Styles and Deviating from Style Normscited by 96
- 2017An Analysis of Visual Question Answering Algorithmscited by 90
- 2021Coordination Among Neural Modules Through a Shared Global Workspacecited by 83
- 2021Neural Production Systems: Learning Rule-Governed Visual Dynamicscited by 75
- 2019Recent Advances in Natural Language Inference: A Survey of Benchmarks, Resources, and Approachescited by 64
12 more
- 2020Dark, Beyond Deep: A Paradigm Shift to Cognitive AI with Humanlike Common Sensecited by 62
- 2018Neural Task Graphs: Generalizing to Unseen Tasks from a Single Video Demonstrationcited by 60
- 2019Learning to Make Analogies by Contrasting Abstract Relational Structurecited by 53
- 2020Learning to Combine Top-Down and Bottom-Up Signals in Recurrent Neural Networks with Attention over Modulescited by 38
- 2018Deep Nets: What have they ever done for Vision?cited by 23
- 2020Going in circles is the way forward: the role of recurrence in visual inferencecited by 19
- 2018A Better Baseline for AVAcited by 16
- 2018The ActivityNet Large-Scale Activity Recognition Challenge 2018 Summarycited by 15
- 2019From explanation to synthesis: Compositional program induction for learning from demonstrationcited by 9
- 2018Adapting Everyday Manipulation Skills to Varied Scenarioscited by 6
- 2018Auto-conditioned Recurrent Mixture Density Networks for Learning Generalizable Robot Skillscited by 2
- 2017Transfer of Tool Affordance and Manipulation Cues with 3D Vision Datacited by 1
Built on by
Citing papers per year: 2022: 1; 2023: 0; 2024: 0; 2025: 0; 2026: 1.
2 papers.