5 papers
Supervised Learning as Lossy Compression: Characterizing Generalization and Sample Complexity via Finite Blocklength Analysis
Kosuke Sugiyama, Masato Uchida
This paper presents a novel information-theoretic perspective on generalization in machine learning by framing the learning problem within the context of lossy compression and appl…
Learning from Similarity-Confidence and Confidence-Difference
Tomoya Tate, Kosuke Sugiyama, Masato Uchida
In practical machine learning applications, it is often challenging to assign accurate labels to data, and increasing the number of labeled instances is often limited. In such case…
Learning from Similarity/Dissimilarity and Pairwise Comparison
Tomoya Tate, Kosuke Sugiyama, Masato Uchida
This paper addresses binary classification in scenarios where obtaining explicit instance level labels is impractical, by exploiting multiple weak labels defined on instance pairs.…
Pool-based Active Learning as Noisy Lossy Compression: Characterizing Label Complexity via Finite Blocklength Analysis
Kosuke Sugiyama, Masato Uchida
This paper proposes an information-theoretic framework for analyzing the theoretical limits of pool-based active learning (AL), in which a subset of instances is selectively labele…
Learning from Hard Labels with Additional Supervision on Non-Hard-Labeled Classes
Kosuke Sugiyama, Masato Uchida
In scenarios where training data is limited due to observation costs or data scarcity, enriching the label information associated with each instance becomes crucial for building hi…