Comments (12)
According to the most recent version, the losses should still be multiplied by 1/2:
Why is a 1/2 missing from your code?
from multi-task-learning-example.
Please have a look at the updated NIPS version of the paper which has some typos corrected.
Yarin
from multi-task-learning-example.
hi @yaringal thanks for your reply
Are you referring to this? I can't seem to find the one with the same title
from multi-task-learning-example.
Ah apologies this is not updated online yet. It will be soon.
The equation above has a type - the implementation is correct.
Yarin
from multi-task-learning-example.
I did the math from scratch and it does seem like there is agreement between code and the most recent version of paper.
Apologies for the trouble.
from multi-task-learning-example.
Here you are https://arxiv.org/pdf/1703.04977.pdf
I am implementing it in pytorch. If you have done it, please share it
from multi-task-learning-example.
@hardianlawi : I reproduced the result from keras using pytorch. YOu can look at https://colab.research.google.com/drive/1_zsmQguerz0iy0J9Uu2Cs7oEHhj0QoXH
However, the sigma2 does not work. Do you have any suggestion @yaringal ?
from multi-task-learning-example.
@John1231983
you need to fix the code
from multi-task-learning-example.
@Banyueqin : Thanks. But this is result after fix it
2627.982177734375
749.9542846679688
538.127685546875
401.61016845703125
308.06640625
241.04234313964844
191.45272827148438
153.8364715576172
124.77854919433594
102.02399444580078
84.0322265625
69.68601989746094
58.185302734375
48.94459533691406
41.494354248046875
35.47389221191406
30.60963249206543
26.680410385131836
23.508365631103516
20.950298309326172
tensor([8.7028]) tensor([5.3088]) 18.911481857299805
from multi-task-learning-example.
from multi-task-learning-example.
No. We expect result is 10 and 0.
from multi-task-learning-example.
@hardianlawi My reuslt goes to negative number after some steps,and I think it results from log_var(sigma),would you mind give me some suggestions?
from multi-task-learning-example.
Related Issues (17)
- some questions about formulation 10 in paper
- uncertainty for self-supervised learning HOT 1
- This is a lucky demo When I change the data generation process, the prediction of the variance is wrong HOT 2
- Question about the loss
- Log var can become negative and explode HOT 1
- Calculating back to actual weights of loss functions
- Why return torch.mean(loss)? HOT 1
- How can I use the trained model for prediction? is it right to use prediction_model.predict(new_x) ? HOT 2
- log_var_a and log_var_b
- Code Doesn't Agree with Paper HOT 1
- about σ HOT 1
- MergeLoss with regular item is \log_{sigma} in paper but \log_{sigma}^2 in code HOT 3
- The loss might be nagative value HOT 11
- Any way to incorporate these methods in other tasks easily? HOT 7
- Question on relative weights
- Sounds like a lucky result comes from a wrong formula deduction HOT 2
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 multi-task-learning-example.