Redundancy-Constrained Information Maximization
Cross-source consensus on Redundancy-Constrained Information Maximization from 1 sources and 5 claims.
1 sources · 5 claims
How it works
Dosage & preparation
Evidence quality
Highlighted claims
- The theoretical objective maximizes mutual information between input signals and tokens while penalizing total correlation among tokens. — Learning Fingerprints for Medical Time Series with Redundancy-Constrained Information Maximization
- The full training loss combines reconstruction and diversity terms with lambda set to 1e-4 in experiments. — Learning Fingerprints for Medical Time Series with Redundancy-Constrained Information Maximization
- The diversity loss uses Total Coding Rate to expand latent volume and encourage orthogonality. — Learning Fingerprints for Medical Time Series with Redundancy-Constrained Information Maximization
- Reconstruction loss is used to maximize a variational lower bound on input-token mutual information. — Learning Fingerprints for Medical Time Series with Redundancy-Constrained Information Maximization
- The paper argues that minimizing the composite loss implements redundancy-constrained information maximization. — Learning Fingerprints for Medical Time Series with Redundancy-Constrained Information Maximization