Published On Mar 2, 2021
6.874/6.802/20.390/20.490/HST.506 Spring 2021 Prof. Manolis Kellis
Deep Learning in the Life Sciences / Computational Systems Biology
Playlist: • MIT Deep Learning in Life Sciences - ...
Latest slides and course today: http://compbio.mit.edu/6874
Spring 2021 slides and materials: http://mit6874.github.io/
0:00 Lecture overview
1:49 CNNs are inspired by the visual cortex
7:57 Biological neurons
14:20 Layers and abstractions
16:45 Features of natural brains
21:05 Illusions
24:20 Key ingredients of a CNN
35:00 Representation learning
39:02 Translating pixels to concepts
40:55 Convolutions
46:20 Kernels: edge detectors and filters
49:05 Representation learning
52:02 Learning a hierarchy of features
54:40 Non-Linearities
56:23 Pooling layers: positional invariance
58:05 Fully connected layers
59:20 Padding, stride, dilation
1:01:30 Data augmentation
1:04:51 Putting it all together
1:10:54 Examples of CNNs
1:13:20 Training CNNs: normalization, initialization, batch size, optimization, tuning
1:20:10 Validation
1:20:38 Summary, Goodbyes