Comments (1)
Hi - that's a good question, we didn't discuss this additional penalty in the paper.
The problem is that with the MSE + non-smooth
loss function, there's a penalty encouraging the first layer's weights to be small, while the next layer's weights aren't penalized at all. That's bad, because you can end up with a model where the first layer's weights are very small, and not necessarily sparse (containing many zeros), but the next layer's weights are very big. So we use the ridge penalty to ensure that the other layer's weights don't get too large.
In our experiments, we tend to fix the ridge penalty to a small value, and only change the non-smooth penalty to find different levels of sparsity. Let me know if this makes sense.
Ian
from neural-gc.
Related Issues (11)
- Code for plotting causal graphs HOT 2
- Regularization Tuning HOT 1
- coding details for a problem similar to DREAM challenge HOT 6
- cMLP: GC_est's weights show 'nan' HOT 1
- Why not get GC and prediction in one model? HOT 2
- GC-est has all zero values HOT 4
- Optimizing lambda HOT 2
- Calculating gradient only on smooth penalty in train_model_ista for cLSTM HOT 1
- NaN loss and all 1 GC HOT 7
- Working with 2-D series HOT 10
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from neural-gc.