Search Results for author:
Found 5 papers, 2 papers with code
Acceleration of Grokking in Learning Arithmetic Operations via Kolmogorov-Arnold Representation
Focusing on the grokking phenomenon that arises in learning arithmetic binary operations via the transformer model, we begin with a discussion on data augmentation in the case of commutative binary operations.
Expressive Power of ReLU and Step Networks under Floating-Point Operations
In this work, we analyze the expressive power of neural networks under a more realistic setup: when we use floating-point numbers and operations.
Self-Knowledge Distillation via Dropout
In this paper, we propose a simple and effective self-knowledge distillation method using a dropout (SD-Dropout).
MCW-Net: Single Image Deraining with Multi-level Connections and Wide Regional Non-local Blocks
Experimental results on both synthetic and real-world rainy datasets demonstrate that the proposed model significantly outperforms existing state-of-the-art models.
Ranked #1 on
Single Image Deraining
on RainCityscapes
Membership Inference Attacks Against Object Detection Models
Machine learning models can leak information regarding the dataset they have trained.
