Efficient Learning of CNNs using Patch Based Features

Alon Brutzkus, Amir Globerson, Eran Malach, Alon Regev Netser, Shai Shalev-Shwartz

نتاج البحث: نشر في مجلةمقالة من مؤنمرمراجعة النظراء


Recent work has demonstrated the effectiveness of using patch based representations when learning from image data. Here we provide theoretical support for this observation, by showing that a simple semi-supervised algorithm that uses patch statistics can efficiently learn labels produced by a one-hidden-layer Convolutional Neural Network (CNN). Since CNNs are known to be computationally hard to learn in the worst case, our analysis holds under some distributional assumptions. We show that these assumptions are necessary and sufficient for our results to hold. We verify that the distributional assumptions hold on real-world data by experimenting on the CIFAR-10 dataset, and find that the analyzed algorithm outperforms a vanilla one-hidden-layer CNN. Finally, we demonstrate that by running the algorithm in a layer-by-layer fashion we can build a deep model which gives further improvements, hinting that this method provides insights about the behavior of deep CNNs.

اللغة الأصليةالإنجليزيّة
الصفحات (من إلى)2336-2356
عدد الصفحات21
دوريةProceedings of Machine Learning Research
مستوى الصوت162
حالة النشرنُشِر - 2022
الحدث39th International Conference on Machine Learning, ICML 2022 - Baltimore, الولايات المتّحدة
المدة: ١٧ يوليو ٢٠٢٢٢٣ يوليو ٢٠٢٢

All Science Journal Classification (ASJC) codes

  • !!Artificial Intelligence
  • !!Software
  • !!Control and Systems Engineering
  • !!Statistics and Probability


أدرس بدقة موضوعات البحث “Efficient Learning of CNNs using Patch Based Features'. فهما يشكلان معًا بصمة فريدة.

قم بذكر هذا