Inhibited Softmax for Uncertainty Estimation in Neural Networks
This provides a lightweight solution for uncertainty estimation in neural networks, beneficial for applications requiring reliability without computational overhead, though it is incremental.
The authors tackled uncertainty estimation and out-of-distribution detection in neural networks by extending the softmax layer with an additional constant input, achieving performance comparable to more expensive methods and outperforming baselines in image recognition and sentiment analysis experiments.
We present a new method for uncertainty estimation and out-of-distribution detection in neural networks with softmax output. We extend softmax layer with an additional constant input. The corresponding additional output is able to represent the uncertainty of the network. The proposed method requires neither additional parameters nor multiple forward passes nor input preprocessing nor out-of-distribution datasets. We show that our method performs comparably to more computationally expensive methods and outperforms baselines on our experiments from image recognition and sentiment analysis domains.