Leveraging Prior Concept Learning Improves Generalization From Few Examples in Computational Models of Human Object Recognition

• Published in: Frontiers in computational neuroscience Vol. 14; p. 586671
• Main Authors: Rule, Joshua S, Riesenhuber, Maximilian
• Format: Journal Article
• Language: English
• Published: Switzerland Frontiers Research Foundation 12.01.2021; Frontiers Media SA; Frontiers Media S.A
• Subjects: artificial neural networks; Computer applications; Concept learning; Deep learning; few-shot learning; Mathematical models; Neural networks; Neuroscience; object recognition; Pattern recognition; semantic cognition; Semantics; transfer learning; Visual system; object recognition; artificial neural networks; semantic cognition; transfer learning; few-shot learning
• ISSN: 1662-5188; 1662-5188
• DOI: 10.3389/fncom.2020.586671

Humans quickly and accurately learn new visual concepts from sparse data, sometimes just a single example. The impressive performance of artificial neural networks which hierarchically pool afferents across scales and positions suggests that the hierarchical organization of the human visual system is critical to its accuracy. These approaches, however, require magnitudes of order more examples than human learners. We used a benchmark deep learning model to show that the hierarchy can also be leveraged to vastly improve the speed of learning. We specifically show how previously learned but broadly tuned conceptual representations can be used to learn visual concepts from as few as two positive examples; reusing visual representations from earlier in the visual hierarchy, as in prior approaches, requires significantly more examples to perform comparably. These results suggest techniques for learning even more efficiently and provide a biologically plausible way to learn new visual concepts from few examples.