Flash presentation by Prof. Gabriel Kreiman in the SciFoo conference 2021
See also:
Bomatter P, Zhang M, Karev D, Madan S, Tseng C, Kreiman G (2021). When Pigs Fly: Contextual Reasoning in Synthetic and Natural Scenes. International Conference on Computer Vision (ICCV) arXiv 2104.02215. PDF
Zhang M, Tseng C, Kreiman G. (2020) Putting visual object recognition in context. CVPR. PDF
Title: Should models of cortex be falsifiable?
Presenters: Prof. Tomaso Poggio (MIT)
Prof. Gabriel Kreiman (Harvard Medical School, BCH)
Prof. Thomas Serre (Brown U.)
Discussants: Prof. Leyla Isik (JHU), Martin Schrimpf (MIT), Michael Lee (MIT), Prof. Susan Epstein (Hunter CUNY), and Jenelle Feather (MIT)
Moderator: Prof. Josh McDermott (MIT)
Date: December 1, 2020 3:00 pm- 5:00 pm
Abstract: Deep Learning architectures designed by engineers and optimized with stochastic gradient descent on large image databases have become de facto models of the cortex. A prominent example is vision. What sorts of insights are derived from these models? Do the performance metrics reveal the inner workings of cortical circuits or are they a dangerous mirage? What are the critical tests that models of cortex should pass?We plan to discuss the promises and pitfalls of deep learning models contrasting them with earlier models (VisNet, HMAX,…) which were developed from the ground up following neuroscience data to account for critical properties of scale + position invariance and selectivity of primate vision.
Lecture presented at the BMM Summer School 2021 by Gabriel Kreiman
Neural Information Processing Systems Review
See also:
Singer JM, Madsen JR, Anderson WS, Kreiman G. (2015). Sensitivity to Timing and Order in Human Visual Cortex. Journal of Neurophysiology, 113:1656-1669.
Singer J, Kreiman G. (2014). Short Temporal Asynchrony Disrupts Visual Object Recognition. Journal of Vision, 12,14
Pattern completion during visual recognition. Gabriel Kreiman. Brown University
See work by Tang et al PNAS 2018
See work by Tang et al Neuron 2014
See review paper about recognition of occluded objects by Hanlin Tang