Training apparatus, classification apparatus, training method, classification method, and program
To provide an efficient and stable training process even in a case where a small amount of target domain labeled data is available. Solution to Problem A training apparatus ( 1 ) comprises: feature extraction means ( 11 ) for extracting source domain feature values from source domain data and for extracting target domain feature values from target domain data; prediction means ( 12 ) for predicting source domain class prediction values from the source domain feature values and for predicting target domain class prediction values from the target domain feature values; filtering means ( 13 ) for computing filtered source domain feature values and filtered target domain feature values by filtering out one or more values from the source domain feature values with reference to the source domain class prediction values and by filtering out one or more values from the target domain feature values with reference to the target domain class prediction values; and updating means ( 14 ) for updating at least one of the feature extraction means and the class prediction means.
1 . A training apparatus comprising at least one processor, the processor carrying out
extraction process for extracting, by one or more feature extractors, source domain feature values from source domain data and for extracting target domain feature values from target domain data;
predicting process for predicting, by one or more class predictors, source domain class prediction values from the source domain feature values and for predicting target domain class prediction values from the target domain feature values;
filtering process for computing filtered source domain feature values and filtered target domain feature values
by filtering out one or more values from the source domain feature values with reference to the source domain class prediction values and
by filtering out one or more values from the target domain feature values with reference to the target domain class prediction values; and
updating process for updating at least one of the one or more feature extractors and one or more class predictors with reference to
a source domain classification loss computed with reference to the source domain class prediction values,
a target domain classification loss computed with reference to the target domain class prediction values, and
a grouping loss computed with reference to the filtered source domain feature values and the filtered target domain feature values.
2 . The training apparatus according to claim 1 , wherein
the filtering process includes
filtering out one or more values from the source domain feature values with reference to the source domain class prediction values and to source domain class label data, and
filtering out one or more values from the target domain feature values with reference to the target domain class prediction values and to target domain class label data.
3 . The training apparatus according to claim 1 , wherein
the filtering process further refer to the source domain classification loss and the target domain classification loss.
4 . The training apparatus according to claim 1 , wherein
the updating process comprises
grouping process for outputting, from the filtered source domain feature values and the filtered target domain feature values, class groups, where each of the class groups contains feature values sharing a same class label.
5 . The training apparatus according to claim 4 , wherein
the updating process further comprises
grouping loss computation process for computing the grouping loss with reference to the class groups.
6 . The training apparatus according to claim 5 , wherein
the grouping loss computation process computes the grouping loss based on
intra class metrics determined with reference to the feature values in a same class, and
inter class metrics determined with reference to the feature values in different classes.
7 . The training apparatus according to claim 1 , wherein
the updating process further comprises
one or more classification loss computation process for computing
the source domain classification loss with reference to the source domain class prediction values and to source domain class label data, and
the target domain classification loss with reference to the target domain class prediction values and to target domain class label data.
8 . The training apparatus according to claim 1 , wherein
the updating process further comprises
merged loss computation means process for computing a merged loss with reference to (i) the source domain classification loss, (ii) the target domain classification loss and (iii) the grouping loss, wherein
the updating process updates at least one of the one or more feature extractors and the one or more class predictors with reference to the merged loss.
9 . The training apparatus according to claim 1 , the processor further carrying out:
one or more domain discrimination process, by a domain discriminator, to discriminate the target domain from the source domain; and
one or more domain loss computation process for outputting a domain discrimination loss as a result of the discrimination process, wherein
the updating process further refers to the domain discrimination loss, and
the updating process further updates the domain discriminator.
10 . A classification apparatus comprising at least one processor, the processor carrying out:
extraction process for extracting, by a feature extractor, target domain feature values from target domain data; and
prediction process for predicting, by a class predictor, target domain class prediction values from the target domain feature values,
wherein
at least one of the feature extractor and the class predictor has been trained with reference to
filtered source domain feature values obtained by filtering out one or more values from source domain feature values, and
filtered target domain feature values obtained by filtering out one or more values from target domain feature values.
11 . A training method comprising:
extracting, by one or more feature extractors, source domain feature values from source domain data and target domain feature values from target domain data;
predicting, by one or more class predictors, source domain class prediction values from the source domain feature values and target domain class prediction values from the target domain feature values;
computing filtered source domain feature values and filtered target domain feature values
by filtering out one or more values from the source domain feature values with reference to the source domain class prediction values and
by filtering out one or more values from the target domain feature values with reference to the target domain class prediction values; and
updating at least one of the one or more feature extractors and the one or more class predictors with reference to
a source domain classification loss computed with reference to the source domain class prediction values,
a target domain classification loss computed with reference to the target domain class prediction values, and
a grouping loss computed with reference to the filtered source domain feature values and the filtered target domain feature values.
12 . A non-transitory storage medium storing therein a program for causing a computer to serve as the training apparatus according to claim 1 , said program causing the computer to carrying out the extraction process, the prediction process, the filtering process, and the updating process.
13 . A non-transitory storage medium storing therein a program for causing a computer to serve as the classification apparatus according to claim 10 , said program causing the computer to carry out the extraction process and the prediction process.