FLEXIBLE MATRIX PROCESSING

1 . A system, comprising:

a matrix transpose component of one or more integrated circuit processors configured to transpose a stored matrix to an output matrix;

a matrix processing component of the one or more integrated circuit processors configured to multiply the output matrix with a mask vector to determine a result vector;

a data modification component of the one or more integrated circuit processors configured to modify at least a portion of the result vector to determine a modified vector; and

a data reduction component of the one or more integrated circuit processors configured to sum at least a portion of elements included in the modified vector.

2 . The system of claim 1 , wherein the stored matrix represents, in an alternative format, an input matrix of values.

3 . The system of claim 2 , wherein at least one element of the stored matrix is represented using a first number of bits.

4 . The system of claim 3 , wherein at least one value of the stored matrix is represented using a second number of bits greater than the first number of bits.

5 . The system of claim 3 , wherein at least one element of the mask vector is a value of one as represented using the first number of bits.

6 . The system of claim 1 , wherein the matrix processing component is configured to multiply the output matrix with the mask vector to determine the result vector including by being configured to compute a dot product of a row of the output matrix with the mask vector.

7 . The system of claim 1 , wherein the data modification component is configured to modify the result vector to determine the modified vector of elements including by being configured to bit shift, by a specified bit shift amount, at least one element of the result vector.

8 . The system of claim 7 , wherein the specified bit shift amount is twenty-four bits, sixteen bits, eight bits, or zero bits.

9 . The system of claim 1 , wherein the matrix transpose component is configured to transpose the stored matrix of elements to output the output matrix including by being configured to copy elements of the stored matrix to a buffer storage.

10 . A method, comprising:

using a matrix transpose component of one or more integrated circuit processors to transpose a stored matrix of elements to an output matrix;

using a matrix processing component of the one or more integrated circuit processors to multiply the output matrix with a mask vector of elements to determine a result vector;

using a data modification component of the one or more integrated circuit processors to modify at least a portion of elements of the result vector to determine a modified vector; and

using a data reduction component of the one or more integrated circuit processors to sum at least a portion of elements of the modified vector.

11 . The method of claim 10 , wherein the stored matrix represents, in an alternative format, an input matrix of values.

12 . The method of claim 11 , wherein at least one element of the stored matrix is represented using a first number of bits.

13 . The method of claim 12 , wherein at least one value of the input matrix is represented using a second number of bits greater than the first number of bits.

14 . The method of claim 11 , further comprising storing the values of the input matrix in a first location of a memory using a specified amount of storage space.

15 . The method of claim 14 , further comprising storing the elements of the stored matrix in a second location of the memory using the specified amount of storage.

16 . The method of claim 11 , wherein the input matrix is utilized in an artificial neural network operation.

17 . The method of claim 12 , wherein at least one element of the mask vector is a value of one as represented using the first number of bits.

18 . The method of claim 10 , wherein using the data modification component of the one or more integrated circuit processors to modify the result vector includes using the one or more integrated circuit processors to bit shift, by a specified bit shift amount, at least one element of the result vector.

19 . The method of claim 18 , wherein the specified bit shift amount is a multiple of eight bits.

20 . A computer program product, the computer program product being embodied in a non-transitory computer readable storage medium and comprising computer instructions for:

using a matrix transpose component to transpose a stored matrix of elements to an output matrix;

using a matrix processing component to multiply the output matrix with a mask vector of elements to determine a result vector;

using a data modification component to modify at least a portion of elements of the result vector to determine a modified vector; and

using a data reduction component to sum at least a portion of elements of the modified vector.