yaringal / multi-task-learning-example

A multi-task learning example for the paper https://arxiv.org/abs/1705.07115

yaringal/multi-task-learning-example Issues

How can I use the trained model for prediction? is it right to use prediction_model.predict(new_x) ?
Updated 7 months ago2
log_var_a and log_var_b
Closed a year ago
Log var can become negative and explode
Updated 2 years ago1
Why return torch.mean(loss)?
Updated 3 years ago1
This is a lucky demo When I change the data generation process, the prediction of the variance is wrong
Updated 3 years ago2
Calculating back to actual weights of loss functions
Closed 3 years ago
Question about the loss
Updated 4 years ago
MergeLoss with regular item is \log_{sigma} in paper but \log_{sigma}^2 in code
Updated 4 years ago3
Sounds like a lucky result comes from a wrong formula deduction
Updated 4 years ago2
uncertainty for self-supervised learning
Updated 4 years ago1
Any way to incorporate these methods in other tasks easily?
Updated 4 years ago7
some questions about formulation 10 in paper
Updated 4 years ago
about σ
Updated 5 years ago1
Question on relative weights
Updated 5 years ago
The loss might be nagative value
Updated 5 years ago11
Final Loss equation
Closed 7 years ago12
Code Doesn't Agree with Paper
Closed 6 years ago1