Introspective Learning : A Two-Stage Approach for Inference in Neural Networks
This addresses the need for more reliable and calibrated neural networks in applications like active learning and out-of-distribution detection, though it is incremental as it builds on existing feed-forward methods.
The paper tackles the problem of improving neural network inference by proposing a two-stage introspective learning approach, which adds a reflection stage to feed-forward decisions, resulting in a 4% increase in robustness and a 42% reduction in calibration errors on noisy data.
In this paper, we advocate for two stages in a neural network's decision making process. The first is the existing feed-forward inference framework where patterns in given data are sensed and associated with previously learned patterns. The second stage is a slower reflection stage where we ask the network to reflect on its feed-forward decision by considering and evaluating all available choices. Together, we term the two stages as introspective learning. We use gradients of trained neural networks as a measurement of this reflection. A simple three-layered Multi Layer Perceptron is used as the second stage that predicts based on all extracted gradient features. We perceptually visualize the post-hoc explanations from both stages to provide a visual grounding to introspection. For the application of recognition, we show that an introspective network is 4% more robust and 42% less prone to calibration errors when generalizing to noisy data. We also illustrate the value of introspective networks in downstream tasks that require generalizability and calibration including active learning, out-of-distribution detection, and uncertainty estimation. Finally, we ground the proposed machine introspection to human introspection for the application of image quality assessment.