Pith. sign in

Extracting Finite Automata from RNNs Using State Merging

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it
abstract

One way to interpret the behavior of a blackbox recurrent neural network (RNN) is to extract from it a more interpretable discrete computational model, like a finite state machine, that captures its behavior. In this work, we propose a new method for extracting finite automata from RNNs inspired by the state merging paradigm from grammatical inference. We demonstrate the effectiveness of our method on the Tomita languages benchmark, where we find that it is able to extract faithful automata from RNNs trained on all languages in the benchmark. We find that extraction performance is aided by the number of data provided during the extraction process, as well as, curiously, whether the RNN model is trained for additional epochs after perfectly learning its target language. We use our method to analyze this phenomenon, finding that training beyond convergence is useful because it leads to compression of the internal state space of the RNN. This finding demonstrates how our method can be used for interpretability and analysis of trained RNN models.

citation-role summary

background 1

citation-polarity summary

fields

cs.LG 1

years

2025 1

verdicts

CONDITIONAL 1

roles

background 1

polarities

background 1

representative citing papers

Learning Model Successors

cs.LG · 2025-01-31 · conditional · novelty 6.0

A meta-learner that transforms a model for difficulty level k into a model for level k+1 can generalize far beyond its training range, demonstrated on a bracket-matching task.

citing papers explorer

Showing 1 of 1 citing paper.

  • Learning Model Successors cs.LG · 2025-01-31 · conditional · none · ref 104 · internal anchor

    A meta-learner that transforms a model for difficulty level k into a model for level k+1 can generalize far beyond its training range, demonstrated on a bracket-matching task.