NEURAL NETWORK LEARNING APPARATUS, NEURAL NETWORK LEARNING METHOD, AND PROGRAM
Provided is a technique for performing learning of a neural network including an encoder and a decoder such that a certain latent variable included in a latent variable vector is larger or the certain latent variable included in the latent variable vector is smaller as a magnitude of a certain property included in an input vector is larger. A neural network learning device performs learning of a neural network including an encoder that converts an input vector into a latent variable vector and a decoder that converts the latent variable vector into an output vector such that the input vector and the output vector are substantially equal to each other, and the learning is performed to cause the latent variable to have monotonicity with respect to the input vector.
1 . A neural network learning device that performs learning of a neural network including an encoder that converts an input vector into a latent variable vector having a latent variable as an element and a decoder that converts the latent variable vector into an output vector such that the input vector and the output vector are substantially identical to each other, wherein
the learning is performed such that,
in a case where two input vectors are set as a first input vector and a second input vector, and a value of an element of the first input vector is greater than a value of an element of the second input vector for at least one element of the input vector, and a value of an element of the first input vector is greater than or equal to a value of an element of the second input vector for all remaining elements of the input vector,
a latent variable vector obtained by converting the first input vector is set as a first latent variable vector, a latent variable vector obtained by converting the second input vector is set as a second latent variable vector, a value of an element of the first latent variable vector is greater than a value of an element of the second latent variable vector for at least one element of the latent variable vector, and a value of an element of the first latent variable vector is greater than or equal to a value of an element of the second latent variable vector for all remaining elements of the latent variable vector.
2 . A neural network learning device that performs learning of a neural network including an encoder that converts an input vector into a latent variable vector having a latent variable as an element and a decoder that converts the latent variable vector into an output vector such that the input vector and the output vector are substantially identical to each other, the neural network learning device comprising
a learning circuitry configured to perform learning by repeating processing of updating parameters of the neural network to cause a value of a loss function to be small, wherein
the loss function includes at least one of
a term having a larger value in a case where a value of a corresponding element of an output vector when the latent variable vector is input is smaller than a value of any element of an output vector when a first artificial latent variable vector is input, the first artificial latent variable vector being a vector in which a value of at least one element of the latent variable vector is replaced with a value smaller than the value, or
a term having a larger value in a case where a value of a corresponding element of an output vector when the latent variable vector is input is larger than a value of any element of an output vector when a second artificial latent variable vector is input, the second artificial latent variable vector being a vector in which a value of at least one element of the latent variable vector is replaced with a value larger than the value.
3 . The neural network learning device according to claim 2 , wherein
a number of an element of the latent variable vector is set as j, and a range of a possible value of a j-th element of the latent variable vector is set as [m j , M j ], where j is an integer greater than or equal to 1 and less than or equal to J, and m j <M j ,
a number of an element of the input vector and the output vector is set as k, and a range of a possible value of a k-th element of the input vector and the output vector is set as [a k , b k ], where k is an integer greater than or equal to 1 and less than or equal to K, K is an integer greater than J, and a k <b k , and
the loss function further includes at least one term of
a value corresponding to a magnitude of a difference between the latent variable vector when the input vector is (b 1 , . . . , b K ) and a vector (M 1 , . . . , M J ),
a value corresponding to a magnitude of a difference between the latent variable vector when the input vector is (a 1 , . . . , a K ) and a vector (m 1 , . . . , m 1 ),
a value corresponding to a magnitude of a difference between the output vector when the latent variable vector is (M 1 , . . . , M J ) and the vector (b 1 , . . . , b K ), or
a value corresponding to a magnitude of a difference between the output vector when the latent variable vector is (m 1 , . . . , m 1 ) and the vector (a 1 , . . . , a K ).
4 . A neural network learning device that performs learning of a neural network including an encoder that converts an input vector into a latent variable vector having a latent variable as an element and a decoder that converts the latent variable vector into an output vector such that the input vector and the output vector are substantially identical to each other, wherein
the learning is performed such that,
in a case where two input vectors are set as a first input vector and a second input vector, and a value of an element of the first input vector is greater than a value of an element of the second input vector for at least one element of the input vector, and a value of an element of the first input vector is greater than or equal to a value of an element of the second input vector for all remaining elements of the input vector,
a latent variable vector obtained by converting the first input vector is set as a first latent variable vector, a latent variable vector obtained by converting the second input vector is set as a second latent variable vector, a value of an element of the first latent variable vector is less than a value of an element of the second latent variable vector for at least one element of the latent variable vector, and a value of an element of the first latent variable vector is less than or equal to a value of an element of the second latent variable vector for all remaining elements of the latent variable vector.
5 . A neural network learning device that performs learning of a neural network including an encoder that converts an input vector into a latent variable vector having a latent variable as an element and a decoder that converts the latent variable vector into an output vector such that the input vector and the output vector are substantially identical to each other, the neural network learning device comprising
a learning circuitry configured to perform learning by repeating processing of updating parameters of the neural network to cause a value of a loss function to be small, wherein
the loss function includes at least one of
a term having a larger value in a case where a value of a corresponding element of an output vector when the latent variable vector is input is larger than a value of any element of an output vector when a first artificial latent variable vector is input, the first artificial latent variable vector being a vector in which a value of at least one element of the latent variable vector is replaced with a value smaller than the value, or
a term having a larger value in a case where a value of a corresponding element of an output vector when the latent variable vector is input is smaller than a value of any element of an output vector when a second artificial latent variable vector is input, the second artificial latent variable vector being a vector in which a value of at least one element of the latent variable vector is replaced with a value larger than the value.
6 . The neural network learning device according to claim 5 , wherein
a number of an element of the latent variable vector is set as j, and a range of a possible value of a j-th element of the latent variable vector is set as [m j , M j ], where j is an integer greater than or equal to 1 and less than or equal to J, and m j <M j ,
a number of an element of the input vector and the output vector is set as k, and a range of a possible value of a k-th element of the input vector and the output vector is set as [a k , b k ], where k is an integer greater than or equal to 1 and less than or equal to K, K is an integer greater than J, and a k <b k , and
the loss function further includes at least one term of
a value corresponding to a magnitude of a difference between the latent variable vector when the input vector is (b 1 , . . . , b K ) and a vector (m 1 , . . . , m J ),
a value corresponding to a magnitude of a difference between the latent variable vector when the input vector is (a 1 , . . . , a K ) and a vector (M 1 , . . . , M J ),
a value corresponding to a magnitude of a difference between the output vector when the latent variable vector is (M 1 , . . . , M J ) and the vector (a 1 , . . . , a K ), or
a value corresponding to a magnitude of a difference between the output vector when the latent variable vector is (m 1 , . . . , m J ) and the vector (b 1 , . . . , b K ).
7 . A neural network learning method in which a neural network learning device performs learning of a neural network including an encoder that converts an input vector into a latent variable vector having a latent variable as an element and a decoder that converts the latent variable vector into an output vector such that the input vector and the output vector are substantially identical to each other, wherein
the learning is performed such that,
in a case where two input vectors are set as a first input vector and a second input vector, and a value of an element of the first input vector is greater than a value of an element of the second input vector for at least one element of the input vector, and a value of an element of the first input vector is greater than or equal to a value of an element of the second input vector for all remaining elements of the input vector,
a latent variable vector obtained by converting the first input vector is set as a first latent variable vector, a latent variable vector obtained by converting the second input vector is set as a second latent variable vector, a value of an element of the first latent variable vector is greater than a value of an element of the second latent variable vector for at least one element of the latent variable vector, and a value of an element of the first latent variable vector is greater than or equal to a value of an element of the second latent variable vector for all remaining elements of the latent variable vector.
8 . A neural network learning method in which a neural network learning device performs learning of a neural network including an encoder that converts an input vector into a latent variable vector having a latent variable as an element and a decoder that converts the latent variable vector into an output vector such that the input vector and the output vector are substantially identical to each other, the neural network learning method comprising
a learning step in which the neural network learning device performs learning by repeating processing of updating parameters of the neural network to cause a value of a loss function to be small, wherein
the loss function includes at least one of
a term having a larger value in a case where a value of a corresponding element of an output vector when the latent variable vector is input is smaller than a value of any element of an output vector when a first artificial latent variable vector is input, the first artificial latent variable vector being a vector in which a value of at least one element of the latent variable vector is replaced with a value smaller than the value, or
a term having a larger value in a case where a value of a corresponding element of an output vector when the latent variable vector is input is larger than a value of any element of an output vector when a second artificial latent variable vector is input, the second artificial latent variable vector being a vector in which a value of at least one element of the latent variable vector is replaced with a value larger than the value.
9 . A neural network learning method in which a neural network learning device performs learning of a neural network including an encoder that converts an input vector into a latent variable vector having a latent variable as an element and a decoder that converts the latent variable vector into an output vector such that the input vector and the output vector are substantially identical to each other, wherein
the learning is performed such that,
in a case where two input vectors are set as a first input vector and a second input vector, and a value of an element of the first input vector is greater than a value of an element of the second input vector for at least one element of the input vector, and a value of an element of the first input vector is greater than or equal to a value of an element of the second input vector for all remaining elements of the input vector,
a latent variable vector obtained by converting the first input vector is set as a first latent variable vector, a latent variable vector obtained by converting the second input vector is set as a second latent variable vector, a value of an element of the first latent variable vector is less than a value of an element of the second latent variable vector for at least one element of the latent variable vector, and a value of an element of the first latent variable vector is less than or equal to a value of an element of the second latent variable vector for all remaining elements of the latent variable vector.
10 . A neural network learning method in which a neural network learning device performs learning of a neural network including an encoder that converts an input vector into a latent variable vector having a latent variable as an element and a decoder that converts the latent variable vector into an output vector such that the input vector and the output vector are substantially identical to each other, the neural network learning method comprising
a learning step in which the neural network learning device performs learning by repeating processing of updating parameters of the neural network to cause a value of a loss function to be small, wherein
the loss function includes at least one of
a term having a larger value in a case where a value of a corresponding element of an output vector when the latent variable vector is input is larger than a value of any element of an output vector when a first artificial latent variable vector is input, the first artificial latent variable vector being a vector in which a value of at least one element of the latent variable vector is replaced with a value smaller than the value, or
a term having a larger value in a case where a value of a corresponding element of an output vector when the latent variable vector is input is smaller than a value of any element of an output vector when a second artificial latent variable vector is input, the second artificial latent variable vector being a vector in which a value of at least one element of the latent variable vector is replaced with a value larger than the value.
11 . A non-transitory recording medium recording a program for causing a computer to function as the neural network learning device according to claim 1 .
12 . A non-transitory recording medium recording a program for causing a computer to function as the neural network learning device according to claim 2 .
13 . A non-transitory recording medium recording a program for causing a computer to function as the neural network learning device according to claim 4 .
14 . A non-transitory recording medium recording a program for causing a computer to function as the neural network learning device according to claim 5 .