[HN Gopher] Understanding the role of individual units in a deep...
       ___________________________________________________________________
        
       Understanding the role of individual units in a deep neural network
        
       Author : Anon84
       Score  : 92 points
       Date   : 2020-12-06 13:30 UTC (1 days ago)
        
 (HTM) web link (www.pnas.org)
 (TXT) w3m dump (www.pnas.org)
        
       | mrfusion wrote:
       | The most interesting use cases for this would be applying it to a
       | game playing AI. We might be able to find novel strategies to
       | games.
       | 
       | Also interesting to find out how the network that solved protein
       | folding works.
        
       | dash2 wrote:
       | This is all super cool, but for those of you who tl:dr;ed, check
       | out their application - creating a "concept painting" app. You
       | can literally click "door" and spray on the picture, and it'll
       | put a door there for you.
       | 
       | http://gandissect.csail.mit.edu
       | 
       | http://gandissect.res.ibm.com/ganpaint.html?project=churchou...
        
       | grantinator wrote:
       | Is this considerably different from this paper? That paper gives
       | the example 'how much does the presence of stripes in this image
       | influence the final classification of zebra?' It seems like the
       | tree example given in the current paper is fairly similar.
       | 
       | https://research.google/pubs/pub47077/
        
       | lamename wrote:
       | Perturbing units and observing change is so much like the early
       | days of neurology and neuroscience. Lesion studies in post-
       | surgery patients or those with stroke or brain damage told us a
       | lot, removing part of the circuit and observing what changed.
        
         | signa11 wrote:
         | yup. i remember reading about something called 'optimal brain
         | damage' which aims at pruning non-essential weights from the
         | network for faster training etc.
        
       | moritzmeister wrote:
       | We also think there should be more Ablation Studies in deep
       | learning research. To make it easier, we are working on
       | automating the process of disabling components of models with a
       | framework called Maggy. Pass in your Keras Model and specify
       | which layes/components to mask out and we will generate all
       | models and train them in parallel.
       | https://databricks.com/session_eu20/parallel-ablation-studie...
       | https://github.com/logicalclocks/maggy
        
       ___________________________________________________________________
       (page generated 2020-12-07 23:01 UTC)