Understanding intermediate layers using linear classifier probes
This provides a diagnostic tool for researchers and practitioners to gain intuition and identify issues in neural network models, though it is incremental as it builds on existing probing techniques.
The paper tackles the problem of understanding neural networks as black boxes by using linear classifier probes to measure feature suitability for classification at each layer, demonstrating that linear separability increases monotonically with depth in models like Inception v3 and ResNet-50.
Neural network models have a reputation for being black boxes. We propose to monitor the features at every layer of a model and measure how suitable they are for classification. We use linear classifiers, which we refer to as "probes", trained entirely independently of the model itself. This helps us better understand the roles and dynamics of the intermediate layers. We demonstrate how this can be used to develop a better intuition about models and to diagnose potential problems. We apply this technique to the popular models Inception v3 and Resnet-50. Among other things, we observe experimentally that the linear separability of features increase monotonically along the depth of the model.