AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Can Neural Networks Understand Logical Entailment?

Richard Evans, David Saxton, David Amos, Pushmeet Kohli, Edward Grefenstette

ICLR 2018 · Architectures and efficiency › RNNs and normalization · approach: RNN training dynamics and gradients · 10 citations (OpenAlex)

Read the paper

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use PossibleWorldNet, evaluating formulas across learned representations of multiple possible worldsPropositional Logic entailment · test (easy)accuracy: 77.8 -> 98.6Improved
Use PossibleWorldNet, evaluating formulas across learned representations of multiple possible worldsPropositional Logic entailment · test (hard)accuracy: 74.2 -> 96.7Improved
Use PossibleWorldNet, evaluating formulas across learned representations of multiple possible worldsPropositional Logic entailment · test (big)accuracy: 74.2 -> 93.9Improved
Use PossibleWorldNet, evaluating formulas across learned representations of multiple possible worldsPropositional Logic entailment · test (massive)accuracy: 59.3 -> 73.4Improved
Use PossibleWorldNet, evaluating formulas across learned representations of multiple possible worldsPropositional Logic entailment · test (exam)accuracy: 85.0 -> 96.0Improved
Use parse-tree structure rather than sequence-only encodingPropositional Logic entailment · test (easy)accuracy: 68.3 -> 77.8Improved
Use convolutional sequence encoders rather than LSTM encodersPropositional Logic entailment · test (easy)accuracy: 68.3 -> 59.7Hurt
Add symbolic vocabulary permutation during training (ablation)Weaker models, including BiDirLSTM encoders and traversal models and convolutional networks · Propositional Logicaccuracy: 2--3 point increaseImproved
Add symbolic vocabulary permutation during training (ablation)Tree-based models · Propositional Logicaccuracy: 7--15 point increasesImproved
Increase the number of imagined possible worlds (ablation)PossibleWorldNet · Propositional Logic validationerror rateImproved
Replace tree-structured formula encoding with LSTM traversal of the joined sequentLSTM Traversal versus LSTM Encoders · Propositional Logic test (easy)accuracy: 68.3 -> 61.8Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy