Differentiable logic networks (DLNs) are interpretable, edge-friendly classifiers that learn both their logic operators and their connections through gradient descent, achieving accuracy competitive with multi-layer perceptrons at orders-of-magnitude lower inference cost. Their accuracy, however, is bounded by the problems of vanishing gradients, scalability, and lack of an ecosystem, leaving a gap relative to conventional neural networks on several datasets. We introduce knowledge distillation for DLNs to narrow this gap. We distill from a trained multi-layer perceptron teacher into a DLN student, integrating the distillation objective into a two-phase training loop without altering the architecture or the inference procedure. The teacher is discarded after training. Hence, the distilled model retains the interpretability, sparsity, and inference cost of a conventionally trained DLN.
We investigate this approach through four variants:
Four Variants Investigated
- A joint hyperparameter search that optimizes the distillation weight and temperature alongside the architecture.
- A grid search over these two hyperparameters with the architecture fixed.
- Two single-phase variants that employ distillation during only the neuron-function phase.
- Two single-phase variants that employ distillation during only the connection phase.
On 20 public tabular classification tasks, distillation improves balanced-class accuracy at negligible cost to gate-level operation counts and model size, though the benefit varies by dataset. Performance improvements of up to 15% are observed, in some cases surpassing the teacher itself.