WisdomNet neural network stops errors by saying when unsure
WisdomNet: trustable machine learning toward error-free classification
Machine LearningArtificial Intelligence
Summary
Machine learning models often make mistakes, even when they're usually right. The authors introduce a way for such models to admit when they aren’t sure about a decision, so they can say 'I don't know' instead of guessing wrong. They created a special type of neural network called WisdomNet that adds a 'conjugate neuron' to detect uncertain cases. This approach lets the model avoid any wrong predictions by rejecting difficult examples and referring them to humans.
What this means in practice
- •For medical imaging teams: Use WisdomNet to produce error-free diagnosis suggestions by deferring uncertain medical images to specialists.
- •For financial fraud detection teams: Implement WisdomNet to flag potentially fraudulent transactions while letting uncertain cases be reviewed by analysts.
Authors
Truong X. Tran, Ramazan S. Aygun
Abstract
Misclassification is a critical problem in many machine learning applications. Since even the classifier models with high accuracy (e.g., > 95%) still introduce some misclassification error, it may not be possible to rely on the output of a classifier. In this paper, we introduce trustable learning, which prompts the learning model to yield only the true output, thus avoiding misclassifications. Whenever the model cannot decide the output accurately, the learning model should indicate that there could be a misclassification error if it is forced to classify, and hence, it should reject to make a decision or defer it to a human expert. Therefore, we develop a methodology for trustable learning and apply it to artificial neural networks and show that it is possible to develop a classifier with 0% misclassification error. We propose a novel neural network architecture named WisdomNet that could provide zero prediction error by introducing an additional neuron named as conjugate neuron that would indicate whether the network is able to classify the data correctly or not. The WisdomNet architecture can be applied to any previously built model, and we have evaluated WisdomNet with several network architectures such as multilayer perceptron, convolutional neural network, and deep network on different data sets. The results show that the WisdomNet is able to reduce the classification error rate to 0%, while labeling the data is difficult to classify as `reject' at a low percentage of within around 10%.