Details
- scope
- Modular LSM (modelling); MNIST benchmark
- claim_text
- A modular liquid state machine with biologically-motivated input synapses outperforms small-world / random LSMs on MNIST and is more robust to damage — relevant to reservoir-computing accounts of cortex.
- section_id
- section_14
- source_url
- https://github.com/AllenNeuralDynamics/ComputationalReviewRecurrence/blob/79ce062d54a924ce05953ec90aa9d26044d2b48f/evidence/section_14_evidence_package.json
- effect_size
- Higher MNIST accuracy and better robustness to damage than small-world / random networks
- review_repo
- ComputationalReviewRecurrence
- section_ref
- wiki_page:computationalreviewrecurrence-14-predictive-coding
- source_kind
- review_finding
- source_path
- evidence/section_14_evidence_package.json
- source_span
- We show that the proposed structure can not only significantly reduce the computational complexity but also achieve higher performance compared to the structure of previous reported networks of a similar size.
- study_system
- Modular LSM (modelling); MNIST benchmark
- section_title
- 14. Predictive-coding and dynamical-systems accounts — the role of recurrent excitatory feedback in error signalling, state estimation, and reservoir computing, evaluated against mouse data
- evidence_summary
- Bio-inspired modular LSM benchmarked on MNIST with Poisson encoding; robustness compared with small-world and random topologies.
- review_bundle_ref
- analysis_bundle:ab-d9c479db9be9
- replication_status
- replication_unknown
- review_package_ref
- analysis_bundle:ab-d9c479db9be9
- source_artifact_ref
- wiki_page:computationalreviewrecurrence-14-predictive-coding
- origin_url
- https://github.com/AllenNeuralDynamics/ComputationalReviewRecurrence/blob/79ce062d54a924ce05953ec90aa9d26044d2b48f/evidence/section_14_evidence_package.json
- commit_sha
- 79ce062d54a924ce05953ec90aa9d26044d2b48f
- created_by
- persona-jerome-lecoq-gbo-neuroscience
- repository_url
- https://github.com/AllenNeuralDynamics/ComputationalReviewRecurrence
Raw fields (4)
- raw_fields
{ "n": null, "doi": "10.3389/fncom.2021.594337", "claim": "A modular liquid state machine with biologically-motivated input synapses outperforms small-world / random LSMs on MNIST and is more robust to damage — relevant to reservoir-computing accounts of cortex.", "cite_key": "Dai2021", "evidence": "Bio-inspired modular LSM benchmarked on MNIST with Poisson encoding; robustness compared with small-world and random topologies.", "effect_size": "Higher MNIST accuracy and better robustness to damage than small-world / random networks", "text_access": "abstract_only", "study_system": "Modular LSM (modelling); MNIST benchmark", "argument_role": "supporting", "replication_status": "replication_unknown", "claim_source_sentence": "We show that the proposed structure can not only significantly reduce the computational complexity but also achieve higher performance compared to the structure of previous reported networks of a similar size.", "source_provenance_status": "non_substring_match", "replication_evidence_dois": [], "effect_size_source_sentence": "We also show that the proposed structure has better robustness against system damage than the small-world and random structures." }- source_refs
[ "paper:paper-61f14ffd1cf7" ]
- evidence_refs
[ { "ref": "paper:paper-61f14ffd1cf7" } ]- source_policy
{ "mode": "public_source_pointer_with_short_context", "notes": [ "Local review repositories are read-only inputs.", "SciDEX stores paper metadata, structured evidence, file pointers, and short citation contexts; it does not copy full review prose." ], "source_commit_sha": "79ce062d54a924ce05953ec90aa9d26044d2b48f", "source_repository_url": "https://github.com/AllenNeuralDynamics/ComputationalReviewRecurrence" }