Convolutional Analysis Operator Learning: Dependence on Training Data
원문 초록
Convolutional analysis operator learning (CAOL) enables the unsupervised training of (hierarchical) convolutional sparsifying operators or autoencoders from large datasets. One can use many training images for CAOL, but a precise understanding of the impact of doing so has remained an open question. This paper presents a series of results that lend insight into the impact of dataset size on the filter update in CAOL. The first result is a general deterministic bound on errors in the estimated filters, and is followed by a bound on the expected errors as the number of training samples increases. The second result provides a high probability analogue. The bounds depend on properties of the training data, and we investigate their empirical values with real data. Taken together, these results provide evidence for the potential benefit of using more training data in CAOL.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
Mesh-Informed Neural Operator : A Transformer Generative Approach
Principled Approaches for Extending Neural Architectures to Function Spaces for Operator Learning
Learning Where to Learn: Training Distribution Selection for Provable OOD Performance
CulFiT: A Fine-grained Cultural-aware LLM Training Paradigm via Multilingual Critique Data Synthesis
Self-Supervised Evolution Operator Learning for High-Dimensional Dynamical Systems
Geometry Aware Operator Transformer as an Efficient and Accurate Neural Surrogate for PDEs on Arbitrary Domains
토론
이 논문과 연결된 포럼 글이 아직 없습니다.