Niels Leadholm on Hierarchical Feature Binding and Robust Machine Vision - September 16th, 2020

I missed this when it came out!

Niels Leadholm, a visiting researcher, discusses his (recently de-anonymized) PhD research on hierarchical feature binding and robust machine vision. He first explores the issue of robust machine vision and his motivation in developing a deep-learning neural network architecture using a biologically-inspired approach. Many AI systems nowadays are vulnerable to adversarial examples. Niels explains how the characteristics of “feature binding,” which happens in a primate’s brain, can be implemented in machine learning systems to enhance robustness.

If you want to follow Niels’ work, you can follow him on Twitter (@neuro_AI).

Papers in the presentation:
Solutions to the Binding Problem - Anne Treisman

Similar binding papers:

The Emergence of Polychronization and Feature Binding in a Spiking Neural Network Model of the Primate Ventral Visual System - Akihiro Eguchi, James B. Isbister, Nasir Ahmad, Simon Stringer

Coding of Border Ownership in Monkey Visual Cortex - Hong Zhou, Howard S. Friedman and Rüdiger von der Heydt

Synchrony unbound: a critical evaluation of the temporal binding hypothesis - M N Shadlen , J A Movshon

Similar papers:

Intriguing properties of neural networks - Christian Szegedy, Wojciech Zaremba, Ilya Sutskever, Joan Bruna, Dumitru Erhan, Ian Goodfellow, Rob Fergus

Scalable Object Detection using Deep Neural Networks - Dumitru Erhan, Christian Szegedy, Alexander Toshev, Dragomir Anguelov

Untangling invariant object recognition - James J DiCarlo, David D Cox

Similar papers:


“Unpooling Layer” is new to me, very clever! Reminds me, loosely, of Hopfield network retrieval.

I wonder if Niels has tried to use attention with Unpooling Layers, similar to attention in the paper “Hopfield Networks is all you need” (you can be hand-wavy and claim hopfield networks are biologically inspired ;))

1 Like