Back

Can Deep Learning Provide a Generalizable Model for Dynamic Sound Encoding in Auditory Cortex?

Pennington, J. R.; David, S. V.

2022-07-22 neuroscience
10.1101/2022.06.10.495698 bioRxiv
Show abstract

Convolutional neural networks (CNNs) can provide powerful and flexible models of neural sensory processing. However, the utility of CNNs in studying the auditory system has been limited by their requirement for large datasets and the complex response properties of single auditory neurons. To address these limitations, we developed a population encoding model: a CNN that simultaneously predicts activity of several hundred neurons recorded during presentation of a large set of natural sounds. This approach defines a shared spectro-temporal space and pools statistical power across neurons. Population models of varying architecture performed consistently better than traditional linear-nonlinear models on data from primary and non-primary auditory cortex. Moreover, population models were highly generalizable. The output layer of a model pre-trained on one population of neurons could be fit to novel single units, achieving performance equivalent to that of neurons in the original fit data. This ability to generalize suggests that population encoding models capture a general set of computations performed by auditory cortex.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

1
Nature Machine Intelligence
○ 70 papers in training set
Top 0.1%
13.1%
Similar papers in this journal
2
PLOS Computational Biology
● 1863 papers in training set
Top 2%
13.1%
Similar papers in this journal
3
Nature Communications
○ 5641 papers in training set
Top 15%
12.1%
Similar papers in this journal
4
Nature Neuroscience
○ 252 papers in training set
Top 0.5%
10.8%
Similar papers in this journal
5
Scientific Reports
○ 3612 papers in training set
Top 6%
8.0%
50% of probability mass above
6
eLife
● 5828 papers in training set
Top 13%
8.0%
Similar papers in this journal
7
Communications Biology
○ 993 papers in training set
Top 1%
5.6%
8
The Journal of Neuroscience
● 1025 papers in training set
Top 5%
3.6%
9
Cell Reports
○ 1498 papers in training set
Top 11%
3.3%
10
Science Advances
● 1243 papers in training set
Top 14%
2.5%
11
Proceedings of the National Academy of Sciences
● 2444 papers in training set
Top 21%
2.4%
12
Frontiers in Computational Neuroscience
○ 60 papers in training set
Top 0.6%
1.9%
13
iScience
○ 1154 papers in training set
Top 20%
1.5%
14
Neuron
○ 337 papers in training set
Top 4%
1.5%
15
Frontiers in Neuroscience
○ 256 papers in training set
Top 4%
1.1%
16
Nature Computational Science
○ 55 papers in training set
Top 1%
1.1%
17
Neural Computation
● 39 papers in training set
Top 0.6%
1.1%
18
PLOS Biology
● 486 papers in training set
Top 11%
0.9%
19
Neural Networks
○ 35 papers in training set
Top 0.6%
0.9%
20
Nature Human Behaviour
○ 95 papers in training set
Top 2%
0.9%
21
Cerebral Cortex
◐ 396 papers in training set
Top 5%
0.6%
22
eneuro
● 439 papers in training set
Top 8%
0.6%

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.