finding

active

finding:gene-ontology-prediction-8-4-auc-improvement-with-unsupervised-autoencoder-and-covariance-pooling-embeddings

Gene Ontology prediction: +8.4% AUC improvement with unsupervised autoencoder and covariance pooling embeddings

Empirical result: covariance pooling combined with unsupervised autoencoder embeddings improves Gene Ontology prediction AUC by 8.4% over mean pooling.

Source paper

extracted_from

Covariance-based Sequence Pooling

(2026) · Dooms, Thomas · Wang, Nicholas K. · Pearce, Michael T.

Neighborhood — ranked by edge-count

Claims (2)

claim

Covariance pooling preserves joint activation structure (feature co-occurrence) that mean pooling discards
supports
Specific interpretive claim about what covariance pooling captures: the pairwise co-activation patterns across features that are invisible to mean pooling.
Second moments preserve structure that first moments destroy.
supports
Core interpretive claim generalizing beyond genomics; argues mean pooling discards information present in covariance.

Communities (3)

community

Manifold-aware concept steering in neural representations
members_of
Explores geometry of activation/behavior manifolds to enable selective, non-destructive concept interventions.
Covariance pooling for high-dimensional genomic embeddings
members_of
Using second-order statistics to compress activation patterns while preserving feature co-occurrence structure, tested on genomic prediction tasks without large labeled datasets.
Covariance pooling for genomic embeddings
members_of
Replaces mean pooling with second-order statistics, achieving large R² and AUC gains on genomic tasks.

Concepts (1)

concept

Gene Ontology Prediction
about
Second evaluated task showing +8.4% AUC improvement with covariance pooling and unsupervised autoencoder embeddings.

Related by similarity (8)

cosine ≥ 0.65 · no typed edge

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

Covariance pooling achieves +52.9% R² improvement over mean pooling on Genomic Track Prediction.finding0.765
Primary empirical result demonstrating practical utility of covariance pooling method.
Natural Language Autoencoders achieve readable explanations through unsupervised reconstruction loss optimized with reinforcement learning, not explicit interpretability constraints.claim0.738
Core insight: reconstruction objective combined with appropriate initialization and KL regularization produces human-interpretable explanations as emergent property.
Gene regulatory networks can evolve associative memory, storing and recalling multiple phenotypes from partial selective cues (Watson et al. 2010)finding0.737
Demonstrates information integration in evolutionary systems with system-level selection
By using a variational autoencoder-like architecture for genomic compression, evolution is freed from over-training and pushed to evolve general-purpose problem-solving machines.claim0.736
Claim linking the indirect genotype-phenotype mapping to robustness and open-endedness.
Sparse autoencoders extract features that are significantly more monosemantic than neurons, as shown by four independent lines of evidenceclaim0.736
Central claim of the paper, supported by detailed feature analysis, human evaluation, automated interpretability of activations, and automated interpretability of logit weights
A/1 autoencoder recovers 79% of MLP log-likelihood loss reduction with 4,096 featuresfinding0.733
Measures how much of the MLP layer's function is explained by the learned features
Gene regulatory network models exhibit associative learning and pattern completion.finding0.732
Analysis of GRN models shows they can perform several kinds of learning, supporting the view of cellular networks as agents on a cognitive continuum.
Sparse autoencoders produce interpretable features for large models.claim0.730
Central claim of the paper: the method scales to state-of-the-art transformers.