A view to a KILL: Knowledge Injection via Lambda Layer

We propose KILL (Knowledge Injection via Lambda Layer) as a novel method for the injection of symbolic knowledge into neural networks (NN) allowing data scientists to control what the network should (not) learn. Unlike other similar approaches, our method does not (i) require ground input formulae, (ii) impose any constraint on the NN undergoing injection, (iii) affect the loss function of the NN. Instead, it acts directly at the backpropagation level, by increasing penalty whenever the NN output is violating the injected knowledge. Experiments are reported to demonstrate the potential (and limits) of our approach.

keywords   symbolic knowledge injection; AI; ML; neural networks; KILL; PSyKI

reference talk

 A view to a KILL: Knowledge Injection via Lambda Layer (WOA 2022, 01/09/2022) — Matteo Magnini (Matteo Magnini, Giovanni Ciatto, Andrea Omicini)

origin event

 WOA 2022

journal or series

 CEUR Workshop Proceedings (CEUR-WS.org)

funding project

 EXPECTATION — Personalized Explainable Artificial Intelligence for decentralized agents with heterogeneous knowledge (01/04/2021–31/03/2024)

reference publication for talk

 A view to a KILL: Knowledge Injection via Lambda Layer (WOA 2022, 01/09/2022) — Matteo Magnini (Matteo Magnini, Giovanni Ciatto, Andrea Omicini)