Learning Explanatory Rules from Noisy Data

Notes

Inductive Logic Programming

Traditional ILP systems are inadequate to deal with noisy, erroneous or ambiguous data. The differentiable approach allows the system to be robust to noise and fuzzy/ambiguous data
The implicit procedure that is learned by a neural network is not inspectable or human-readable. (Why? Need to understand what author meant)
Performance of systems diminishes when the test data is significantly larger than training data.

“If we train the neural system to add numbers of length 10, they may also be successful when tested on numbers of length 20. But if we test them on numbers of length 100, the performance deteriorates as shown in Kaiser, 2015; Reed & de Freitas, 2015