/* ---- Google Analytics Code Below */

Monday, April 29, 2019

Feedback loops for Neural Nets

I remember something similar posed in the 90s, tat feedback could converge to better results. Or at least produce some better next step position.   Don't recall any work suggest progress back then.  Now its here?  Lots more detail at the link.

For Better Deep Neural Network Vision, just add Feedback (loops)  in MIT News
The DiCarlo lab finds that a recurrent architecture helps both artificial intelligence and our brains to better identify objects.

Sabbi Lall | McGovern Institute for Brain Research 

Your ability to recognize objects is remarkable. If you see a cup under unusual lighting or from unexpected directions, there’s a good chance that your brain will still compute that it is a cup. Such precise object recognition is one holy grail for artificial intelligence developers, such as those improving self-driving car navigation.

While modeling primate object recognition in the visual cortex has revolutionized artificial visual recognition systems, current deep learning systems are simplified, and fail to recognize some objects that are child’s play for primates such as humans.

In findings published in Nature Neuroscience, McGovern Institute investigator James DiCarlo and colleagues have found evidence that feedback improves recognition of hard-to-recognize objects in the primate brain, and that adding feedback circuitry also improves the performance of artificial neural network systems used for vision applications. ... " 

More on Dicarlo Lab at MIT, with more on the above and related activity

No comments: