AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

What Makes Data Suitable for a Locally Connected Neural Network? A Necessary and Sufficient Condition Based on Quantum Entanglement.

Yotam Alexander, Nimrod De La Vega, Noam Razin, Nadav Cohen

NeurIPS 2023 · Architectures and efficiency › RNNs and normalization · approach: Theory of depth and width · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

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
Rearrange randomly permuted features with the entanglement-based algorithmCNN · Speech Commands (randomly permuted)test accuracy: 5.5 -> 18Improved
Rearrange randomly permuted features with the entanglement-based algorithmS4 · Speech Commands (randomly permuted)test accuracy: 9.4 -> 30.1Improved
Rearrange randomly permuted features with the entanglement-based algorithmLocal-Attention · Speech Commands (randomly permuted)test accuracy: 7.8 -> 12.9Improved
Rearrange features with IGTD instead of random permutationCNN · Speech Commands (randomly permuted)test accuracy: 5.5 -> 5.7No clear effect
Rearrange features with IGTD instead of random permutationS4 · Speech Commands (randomly permuted)test accuracy: 9.4 -> 12.8Improved
Use entanglement-based arrangement instead of IGTDS4 · dnatest accuracy: 89.9 -> 89.1Hurt
Rearrange tabular features with the entanglement-based algorithmCNN · dnatest accuracy: 82.5 -> 91.2Improved
Rearrange tabular features with the entanglement-based algorithmLocal-Attention · semeiontest accuracy: 60.9 -> 78.0Improved
Rearrange tabular features with the entanglement-based algorithmS4 · isolettest accuracy: 92.3 -> 93.4Improved
Rearrange tabular features with the entanglement-based algorithmCNN · semeiontest accuracy: 77.7 -> 80.0Improved
Increase number of random feature swaps in audio dataCNN, S4, Local-Attention · Speech Commands binaryentanglement and test accuracyHurt
Rearranged features vs original arrangement on Speech CommandsCNN · Speech Commandstest accuracy: 59.8 -> 18Hurt

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