Skip to main content
12 votes
Accepted

Multi task learning in Keras

By using the functional API you can easily share weights between different parts of your network. In your case we have an Input $x$ which is our input, then we will ...
Jan van der Vegt's user avatar
4 votes
Accepted

Missing outputs in multiple-output neural net

Since we are talking about multiple different types of targets (classes versus numerical for example) we already need a composite loss function. I will consider how to balance the different composite ...
Jan van der Vegt's user avatar
4 votes

Multi target classification for different types of target variables

You have one classification task and one regression task, but sklearn's multioutput meta-estimators only support two tasks of the same type. The best solution here is to train two models: A binary ...
Imran's user avatar
  • 2,381
3 votes
Accepted

Multi task learning architecture for Multi-label classification

You should design a multi-task model (MTM). MTM has the ability to share learned representations from input between several tasks. More precisely, we try to simultaneously optimize a model with ...
Mo-'s user avatar
  • 1,255
2 votes

Multi target classification for different types of target variables

You should break this down into two models. I would solve this in the following manner: The first model would predict if its either Target 1 or Target 2 by looking at 100 columns The second model ...
Nischal Hp's user avatar
2 votes
Accepted

Should I rescale losses before combining them for multitask learning?

Check this. Under the heading Multi-tasks losses they have mentioned how they average losses from two different tasks. They do a weighted average depending on their use case.
ashutosh singh's user avatar
2 votes

Implementing Multitask Classification in Keras

You need to pass an array object. You don't need to map the label array with the output node name ( like TensorFlow low-level APIs ). Pass the list containing the label arrays directly. ...
Shubham Panchal's user avatar
2 votes

Emotion Recognition with Multi-task Learning

I suppose your question is .... "How to do multi-task learning?" Should you take a neural network approach, for instance: A simple example with the above diagram is to pass Columns 1-8 into ...
David Tang's user avatar
2 votes
Accepted

How to feed data to multi-output Keras model from a single TFRecords file

After playing around with tf.data.map operations I found the answer was easier than expected, I simply had to preprocess the data and put all the labels for each ...
magomar's user avatar
  • 131
1 vote

How to compute sample weight in multi-task model?

In keras, there is a class_weight argument. Create a dictionary of weights per class and then pass that into the .fit method of the model. ...
Brian Spiering's user avatar
1 vote

Multi-modal neural network

multimodal learning can be complex (like anything) but it can also be fairly simple. The general idea of multimodal modeling is to take data consumed in parallel, which has different "modes" ...
Warlax56's user avatar
  • 430
1 vote
Accepted

What is the difference between Multi task learning and domain generalization

Domain generalization: Aims to train a model using multi-domain source data, such that it can directly generalize to new domains without need of retraining. Focusing, Multiple domains on same task ...
Archana David's user avatar
1 vote

Combining heterogeneous data sets for more powerful machine learning

Following your considerations: IMDB and RT don't have all the same movies I would build a larger dataset of movies, aggregate all observations available, and run a model on that. There are ...
Leevo's user avatar
  • 6,265
1 vote

Architecture for multivariate multi-time-series model where some features are TS specific and some features are global

Stacked LSTM is one option in this scenario This assumes that First two LSTMs have different frequencies and City has static features (Like lat/long, one-hot-encoded value etc). If City is also time-...
Shamit Verma's user avatar
  • 2,259
1 vote

Thoughts on improving the Multitask Learning Model

Adjust the optimizer hyper-parameters 1) use amsgrad = True 2) try lower beta_1 3) adjust lr from 1e-6 to .1 or use a learning rate finder 4) use either the weight_decay(learning schedule) in the ...
Kyle V.'s user avatar
  • 11
1 vote

Multitask learning NN only trains on a few tasks

So, it turns out that my problem had a lot more to do with network structure than anything else. I was trying to share too many lower levels, and then still not leaving enough room for the later ...
THerzog's user avatar
  • 31
1 vote

Multi-task learning for Multi-label classification?

MTL is supposed to enable your shared layer to generalise better.For eg. in a text classification problem your encoder would be ...
ashutosh singh's user avatar
1 vote

Multi-Source Time Series Data Prediction

The answer is Data Fusion or Feature Fusion. I am implementing one using neural networks to classify Human Activity using multiple kinds of sensors: accelerometer, binary sensors, ecc.. We train a ...
Francesco Pegoraro's user avatar
1 vote

Multi-Source Time Series Data Prediction

You can certainly use an LSTM for this approach, as well as VAR, or potentially a more 'traditional' model (random forest). I have implemented multivariate LSTM for t+x prediction in Keras. Feature ...
Hobbes's user avatar
  • 1,459
1 vote

Multi-Source Time Series Data Prediction

I'm only aware (and using) a RNN which gets multiple time-series in it's first layer and then mixes those in the following layers. Let me know if you have a question on this approach.
Tobi's user avatar
  • 166
1 vote

Error on multitask neural nets where all outputs not observed for every example

You have already found the answer by yourself: if there are missing values, you should not backpropagate on them. I do not know of any particular name for this action. I do not think that there are ...
CharlesG's user avatar
  • 267

Only top scored, non community-wiki answers of a minimum length are eligible