Augmented gamma belief network operation

1. A computer-implementable method for generating a cognitive insight comprising:

receiving data, the data comprising a plurality of examples, each of the plurality of examples comprising an input object and a desired output value, at least some of the plurality of examples being based upon feedback from a user;

performing a machine learning operation on the data, the machine learning operation comprising performing an augmented gamma belief network operation, the augmented gamma belief network operation producing an inferred function based upon the data;

performing a cognitive learning operation via a cognitive inference and learning system using the plurality of examples, the cognitive learning operation implementing a cognitive learning technique according to a cognitive learning framework, the cognitive learning framework comprising a plurality of cognitive learning styles and a plurality of cognitive learning categories, each of the plurality of cognitive learning styles comprising a generalized learning approach implemented by the cognitive inference and learning system to perform the cognitive learning operation, each of the plurality of cognitive learning categories referring to a source of information used by the cognitive inference and learning system when performing the cognitive learning operation, an individual cognitive learning technique being associated with a primary cognitive learning style and bounded by an associated primary cognitive learning category, the cognitive learning operation applying the cognitive learning technique via the machine learning operation to generate a cognitive learning result;

generating a cognitive insight based upon a cognitive profile generated using the inferred function generated by the augmented gamma belief network operation; and wherein

the plurality of cognitive learning techniques comprising a direct correlations cognitive learning technique, an explicit likes/dislikes cognitive learning technique, a patterns and concepts cognitive learning technique, a behavior cognitive learning technique, a concept entailment cognitive learning technique, and a contextual recommendation cognitive learning technique, the direct correlations cognitive learning technique being associated with a declared learning style and bounded by a data-based cognitive learning category, an explicit likes/dislikes cognitive learning technique being associated with the declared learning style and bounded by an interaction-based cognitive learning category, the patterns and concepts cognitive learning technique being associated with an observed learning style and bounded by the data-based cognitive learning category, the behavior cognitive learning technique being associated with the observed learning style and bounded by the interaction-based cognitive learning category, the concept entailment cognitive learning technique being associated with an inferred learning style and bounded by the data-based cognitive learning category, and a contextual recommendation cognitive learning technique being associated with the inferred learning style and bounded by the interaction-based cognitive learning category.

2. The method of claim 1 , wherein:

the augmented gamma belief network operation factorizes each of a plurality of hidden layers into a product of a space connection weight matrix.

3. The method of claim 1 , wherein:

the augmented gamma belief network operation factorizes each of a plurality of hidden layers nonnegative real hidden units of a next layer of abstraction.

4. The method of claim 1 , wherein:

each of a plurality of hidden layers are trained via a Gibbs sampler operation, the Gibb sampler operation performing an upward sampling operation and a downward sampling operation on each of the plurality of hidden layers.

5. The method of claim 4 , wherein: each upward sampling operation propagates latent counts and samples Dirichlet distributed connection weight vectors starting from a bottom-most layer of the plurality of hidden layers.

6. The method of claim 5 , wherein:

each downward sampling operation samples gamma distributed hidden units starting from a top hidden layer with each next lower hidden layer solved with a same subroutine.

7. A system comprising:

a processor;

a data bus coupled to the processor; and

a non-transitory, computer-readable storage medium embodying computer program code, the non-transitory, computer-readable storage medium being coupled to the data bus, the computer program code interacting with a plurality of computer operations and comprising instructions executable by the processor and configured for:

receiving data, the data comprising a plurality of examples, each of the plurality of examples comprising an input object and a desired output value, at least some of the plurality of examples being based upon feedback from a user;

performing a machine learning operation on the data, the machine learning operation comprising performing an augmented gamma belief network operation, the augmented gamma belief network operation producing an inferred function based upon the data;

performing a cognitive learning operation via a cognitive inference and learning system using the plurality of examples, the cognitive learning operation implementing a cognitive learning technique according to a cognitive learning framework, the cognitive learning framework comprising a plurality of cognitive learning styles and a plurality of cognitive learning categories, each of the plurality of cognitive learning styles comprising a generalized learning approach implemented by the cognitive inference and learning system to perform the cognitive learning operation, each of the plurality of cognitive learning categories referring to a source of information used by the cognitive inference and learning system when performing the cognitive learning operation, an individual cognitive learning technique being associated with a primary cognitive learning style and bounded by an associated primary cognitive learning category, the cognitive learning operation applying the cognitive learning technique via the machine learning operation to generate a cognitive learning result;

generating a cognitive insight based upon a cognitive profile generated using the inferred function generated by the augmented gamma belief network operation; and wherein

the plurality of cognitive learning techniques comprising a direct correlations cognitive learning technique, an explicit likes/dislikes cognitive learning technique, a patterns and concepts cognitive learning technique, a behavior cognitive learning technique, a concept entailment cognitive learning technique, and a contextual recommendation cognitive learning technique, the direct correlations cognitive learning technique being associated with a declared learning style and bounded by a data-based cognitive learning category, an explicit likes/dislikes cognitive learning technique being associated with the declared learning style and bounded by an interaction-based cognitive learning category, the patterns and concepts cognitive learning technique being associated with an observed learning style and bounded by the data-based cognitive learning category, the behavior cognitive learning technique being associated with the observed learning style and bounded by the interaction-based cognitive learning category, the concept entailment cognitive learning technique being associated with an inferred learning style and bounded by the data-based cognitive learning category, and a contextual recommendation cognitive learning technique being associated with the inferred learning style and bounded by the interaction-based cognitive learning category.

8. The system of claim 7 , wherein:

the augmented gamma belief network operation factorizes each of a plurality of hidden layers into a product of a space connection weight matrix.

9. The system of claim 7 , wherein:

the augmented gamma belief network operation factorizes each of a plurality of hidden layers nonnegative real hidden units of a next layer of abstraction.

10. The system of claim 7 , wherein:

each of a plurality of hidden layers are trained via a Gibbs sampler operation, the Gibb sampler operation performing an upward sampling operation and a downward sampling operation on each of the plurality of hidden layers.

11. The system of claim 10 , wherein:

each upward sampling operation propagates latent counts and samples Dirichlet distributed connection weight vectors starting from a bottom-most layer of the plurality of hidden layers.

12. The system of claim 10 , wherein:

each downward sampling operation samples gamma distributed hidden units starting from a top hidden layer with each next lower hidden layer solved with a same subroutine.

13. A non-transitory, computer-readable storage medium embodying computer program code, the computer program code comprising computer executable instructions configured for:

receiving data, the data comprising a plurality of examples, each of the plurality of examples comprising an input object and a desired output value, at least some of the plurality of examples being based upon feedback from a user;

performing a machine learning operation on the data, the machine learning operation comprising performing an augmented gamma belief network operation, the augmented gamma belief network operation producing an inferred function based upon the data;

performing a cognitive learning operation via a cognitive inference and learning system using the plurality of examples, the cognitive learning operation implementing a cognitive learning technique according to a cognitive learning framework, the cognitive learning framework comprising a plurality of cognitive learning styles and a plurality of cognitive learning categories, each of the plurality of cognitive learning styles comprising a generalized learning approach implemented by the cognitive inference and learning system to perform the cognitive learning operation, each of the plurality of cognitive learning categories referring to a source of information used by the cognitive inference and learning system when performing the cognitive learning operation, an individual cognitive learning technique being associated with a primary cognitive learning style and bounded by an associated primary cognitive learning category, the cognitive learning operation applying the cognitive learning technique via the machine learning operation to generate a cognitive learning result;

generating a cognitive insight based upon a cognitive profile generated using the inferred function generated by the augmented gamma belief network operation; and wherein

the plurality of cognitive learning techniques comprising a direct correlations cognitive learning technique, an explicit likes/dislikes cognitive learning technique, a patterns and concepts cognitive learning technique, a behavior cognitive learning technique, a concept entailment cognitive learning technique, and a contextual recommendation cognitive learning technique, the direct correlations cognitive learning technique being associated with a declared learning style and bounded by a data-based cognitive learning category, an explicit likes/dislikes cognitive learning technique being associated with the declared learning style and bounded by an interaction-based cognitive learning category, the patterns and concepts cognitive learning technique being associated with an observed learning style and bounded by the data-based cognitive learning category, the behavior cognitive learning technique being associated with the observed learning style and bounded by the interaction-based cognitive learning category, the concept entailment cognitive learning technique being associated with an inferred learning style and bounded by the data-based cognitive learning category, and a contextual recommendation cognitive learning technique being associated with the inferred learning style and bounded by the interaction-based cognitive learning category.

14. The non-transitory, computer-readable storage medium of claim 13 , wherein:

the augmented gamma belief network operation factorizes each of a plurality of hidden layers into a product of a space connection weight matrix.

15. The non-transitory, computer-readable storage medium of claim 13 , wherein:

the augmented gamma belief network operation factorizes each of a plurality of hidden layers nonnegative real hidden units of a next layer of abstraction.

16. The non-transitory, computer-readable storage medium of claim 13 , wherein:

each of a plurality of hidden layers are trained via a Gibbs sampler operation, the Gibb sampler operation performing an upward sampling operation and a downward sampling operation on each of the plurality of hidden layers.

17. The non-transitory, computer-readable storage medium of claim 16 , wherein:

each upward sampling operation propagates latent counts and samples Dirichlet distributed connection weight vectors starting from a bottom-most layer of the plurality of hidden layers.

18. The non-transitory, computer-readable storage medium of claim 16 , wherein:

each downward sampling operation samples gamma distributed hidden units starting from a top hidden layer with each next lower hidden layer solved with a same subroutine.

19. The non-transitory, computer-readable storage medium of claim 13 , wherein the computer executable instructions are deployable to a client system from a server system at a remote location.

20. The non-transitory, computer-readable storage medium of claim 13 , wherein the computer executable instructions are provided by a service provider to a user on an on-demand basis.