TY - GEN
T1 - Deep Active Learning with Simulated Rationales for Text Classification
AU - Guélorget, Paul
AU - Grilheres, Bruno
AU - Zaharia, Titus
N1 - Publisher Copyright:
© 2020, Springer Nature Switzerland AG.
PY - 2020/1/1
Y1 - 2020/1/1
N2 - Neural networks have become a preferred tool for text classification tasks, demonstrating state of the art performances when trained on a large set of labeled data. However, in an early active learning setup, the scarcity of the ground-truth labels available severely penalizes the generalization capability of the neural network. In order to overcome such limitations, in this paper, we introduce a new learning strategy, which consist of inserting in the early stages of the learning process some additional, local and salient knowledge, presented under the form of simulated, human like rationales. We show how such knowledge can be automatically extracted from documents by analyzing the class activation maps of a convolutional neural network. The experimental results obtained demonstrate that the exploitation of such rationales permits to significantly speed-up the learning process, with a spectacular increase of the accuracy rates, starting from a very reduced number of documents (10–20).
AB - Neural networks have become a preferred tool for text classification tasks, demonstrating state of the art performances when trained on a large set of labeled data. However, in an early active learning setup, the scarcity of the ground-truth labels available severely penalizes the generalization capability of the neural network. In order to overcome such limitations, in this paper, we introduce a new learning strategy, which consist of inserting in the early stages of the learning process some additional, local and salient knowledge, presented under the form of simulated, human like rationales. We show how such knowledge can be automatically extracted from documents by analyzing the class activation maps of a convolutional neural network. The experimental results obtained demonstrate that the exploitation of such rationales permits to significantly speed-up the learning process, with a spectacular increase of the accuracy rates, starting from a very reduced number of documents (10–20).
KW - Active learning
KW - Class activation maps
KW - Deep neural networks
KW - Rationales
KW - Text classification
U2 - 10.1007/978-3-030-59830-3_32
DO - 10.1007/978-3-030-59830-3_32
M3 - Conference contribution
AN - SCOPUS:85092893707
SN - 9783030598297
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 363
EP - 379
BT - Pattern Recognition and Artificial Intelligence - International Conference, ICPRAI 2020, Proceedings
A2 - Lu, Yue
A2 - Vincent, Nicole
A2 - Yuen, Pong Chi
A2 - Zheng, Wei-Shi
A2 - Cheriet, Farida
A2 - Suen, Ching Y.
PB - Springer Science and Business Media Deutschland GmbH
T2 - 2nd International Conference on Pattern Recognition and Artificial Intelligence, ICPRAI 2020
Y2 - 19 October 2020 through 23 October 2020
ER -