Publications by authors named "Jean-Nicolas Jeremie"
Article Synopsis
- Humans are naturally skilled at quickly categorizing images, especially when identifying animals, while deep learning algorithms using convolutional neural networks (CNNs) have recently surpassed human accuracy in specific visual tasks.
- Despite the high accuracy of CNNs, they struggle with generalization, such as recognizing objects after image rotations, making biological visual systems more adaptable for broader tasks.
- The study retrained the VGG 16 CNN for two ecologically relevant tasks (detecting animals or artifacts) and showed it could achieve human-like performance, demonstrating that fewer layers in CNNs can still yield strong results, which could inform both artificial intelligence designs and psychophysical research.
Why do neurons communicate through spikes? By definition, spikes are all-or-none neural events which occur at continuous times. In other words, spikes are on one side binary, existing or not without further details, and on the other, can occur at any asynchronous time, without the need for a centralized clock. This stands in stark contrast to the analog representation of values and the discretized timing classically used in digital processing and at the base of modern-day neural networks.
View Article and Find Full Text PDF