Multi-task equidistant embedding

1. A method comprising:

receiving, by at least one processing device, a dataset corresponding to a plurality of tasks, the dataset describing a dimension space having a plurality of categories and a plurality of features formatted as one-hot encoding vectors corresponding to tasks included within respective categories of the plurality of categories, the plurality of features having equidistant relationships, one to another, within respective said categories in the dimension space;

generating, by the at least one processing device, a shared representation of the plurality of the tasks of the dataset based on a reduced dimension dataset in a reduced dimension space that is reduced in comparison with the dimension space of the dataset, the reduced dimension dataset maintaining the equidistant relationships, one to another, of the plurality of features, respectively, within the respective said categories from the dataset by:

applying constraints to the dataset using an embedding loss function based on a comparison between a dense embedding matrix and a binary feature matrix;

concatenating embeddings of the reduced dimension dataset;

performing batch normalization on the reduced dimension dataset; and

reducing overfitting of the reduced dimension dataset by randomly dropping units from the reduced dimension dataset;

generating, by the at least one processing device, an exclusive representation of a particular task of the dataset by extracting complementary information including feature interactions by:

segmenting the particular task from the plurality of tasks of the dataset using k-means clustering; and

comparing the particular task to the plurality of tasks of the dataset using a joint loss function that includes an embedding loss and a segmentation loss based on the k-means clustering; and

training, by the at least one processing device, a machine learning model based on the reduced dimension dataset of the shared representation and the complementary information of the exclusive representation to generate a prediction of an outcome of the particular task of the dataset from the exclusive representation.

2. The method of claim 1 , further comprising:

receiving, by the at least one processing device, an additional dataset that corresponds to the particular task, the additional dataset describing the plurality of categories and the plurality of features;

generating, by the least one processing device, a prediction of an outcome of the particular task based on the additional dataset and the trained machine learning model; and

outputting, by the at least one processing device, the generated prediction of the outcome of the particular task.

3. The method of claim 1 , wherein the plurality of tasks comprises at least a first task and a second task, and wherein the training the machine learning model comprises training a first machine learning model corresponding to the first task and training a second machine learning model corresponding to the second task.

4. The method of claim 3 , wherein the first task is a supervised task and the second task is one of a supervised task or an unsupervised task.

5. The method of claim 3 , wherein the first machine learning model and the second machine learning model are trained from a neural network that includes shared layers corresponding to the first task and the second task, exclusive layers corresponding to the first task, and exclusive layers corresponding to the second task.

6. The method of claim 3 , wherein the second machine learning model is trained utilizing information corresponding to the first task.

7. The method of claim 1 , wherein the reduced dimension dataset includes an equidistant embedding that enforces, for each respective category of the plurality of categories, the equidistant relationship among the plurality of features included within the respective category.

8. The method of claim 1 , wherein the generating the reduced dimension dataset is performed as a part of the training the machine learning model.

9. At least one processing device including a processing system and at least one computer-readable storage medium, the at least one processing device comprising:

equidistant embedding layers of a neural network, the equidistant embedding layers configured to enforce, for each respective category of a plurality of categories described in a dataset, an equidistant relationship defined using a distance measurement among a plurality of features formatted as one-hot encoding vectors corresponding to tasks included within the respective category in a reduced dimension dataset in a reduced dimension space from a dimension space that is maintained from the dataset, the reduced dimension space generated by:

applying constraints to the dataset using an embedding loss function based on a comparison between a dense embedding matrix and a binary feature matrix;

concatenating embeddings of the reduced embedding space;

performing batch normalization on the reduced dimension dataset; and

reducing overfitting of the reduced dimension dataset by randomly dropping units from the reduced dimension dataset;

shared layers of the neural network, the shared layers configured to extract feature interactions between the features described in the dataset corresponding to at least one of a first task and a second task;

exclusive layers of the neural network corresponding to the first task, the exclusive layers corresponding to the first task configured to utilize the extracted feature interactions to generate a first machine learning model corresponding to the first task by:

segmenting the first task from the plurality of tasks of the dataset using k-means clustering; and

comparing the first task to the plurality of tasks of the dataset using a joint loss function that includes an embedding loss and a segmentation loss based on the k-means clustering; and

exclusive layers of the neural network corresponding to the second task, the exclusive layers corresponding to the second task configured to utilize the extracted feature interactions to generate a second machine learning model corresponding to the second task.

10. The at least one processing device of claim 9 , the at least one computer-readable storage medium storing processor-executable instructions that, responsive to execution by the processing system, cause the processing system to perform operations comprising:

receiving a second dataset corresponding to the first task;

generating a prediction of an outcome of the first task based on the second dataset and the first machine learning model; and

outputting the generated prediction of the outcome of the first task.

11. The at least one processing device of claim 9 , wherein the first machine learning model and the second machine learning model are generated concurrently.

12. The at least one processing device of claim 9 , wherein the first task is a supervised task and the second task is an unsupervised task.

13. The at least one processing device of claim 9 , wherein the first task is a supervised task and the second task is a supervised task.

14. The at least one processing device of claim 9 , wherein the extracting feature interactions includes determining complementary information beneficial to the first task and the second task.

15. The at least one processing device of claim 9 , wherein the extracting feature interactions includes determining a feature interaction beneficial to the first task, and wherein the generating the second machine learning model corresponding to the second task includes utilizing the determined feature interaction.

16. The at least one processing device of claim 9 , wherein the generating the first machine learning model and the generating the second machine learning model are subject to same training criterion.

17. A computer-readable storage medium storing executable instructions, which when executed by a processing device, cause the processing device to perform operations comprising:

receiving a dataset corresponding to at least one task, the first dataset describing a three-dimensional Euclidean space having three categories and a plurality of features formatted as one-hot encoding vectors corresponding to tasks included within respective said categories, the plurality of features having equidistant relationships, one to another, within respective said categories in the three-dimensional space;

generating training data that enforces the equidistant relationships of the plurality of features corresponding to the dataset in the three-dimensional space within respective said categories in a two-dimensional space by:

applying constraints to the dataset using an embedding loss function based on a comparison between a dense embedding matrix and a binary feature matrix;

extracting complementary information including feature interactions by:

segmenting a particular feature from the plurality of features of the dataset using k-means clustering; and

comparing the particular feature to the plurality of features of the dataset using a joint loss function that includes an embedding loss and a segmentation loss based on the k-means clustering;

concatenating the training data with the complementary information;

training a machine learning model using the training data;

performing batch normalization on the training data; and

reducing overfitting of the training data by randomly dropping units from the training data;

receiving an input corresponding to the at least one task;

generating a prediction of an outcome of the at least one task based on the input and the trained machine learning model; and

outputting the generated prediction of the outcome of the at least one task.

18. The computer-readable storage medium of claim 17 , wherein the at least one task comprises at least a first task and a second task, and wherein training the machine learning model comprises training a first machine learning model corresponding to the first task and means for training a second machine learning model corresponding to the second task.

19. The computer-readable storage medium of claim 18 , wherein the first task is a supervised task and the second task is an unsupervised task.

20. The computer-readable storage medium of claim 18 , wherein the first task and the second task are unsupervised tasks.