Computer Science
4 years ago

New AI model learns to see the world better than humans.

2 views
Learning Multiple Layers of Features from Tiny Images
Alex Krizhevsky

Paper Summary

The article explains how a group trained a computer model to recognize patterns in small color images. They used RBMs and DBNs to learn filters that help a computer tell objects apart better than just looking at the pixels. They trained the model on a set of labeled images called CIFAR-10.

Related papers

Theme background image
Computer Science
5 years ago

Neural networks can be tricked into misclassifying images with imperceptible changes.

Neural Networks
Adversarial Robustness
Anomaly Detection
Theme background image
Computer Science
5 years ago

New Optimization Algorithm Adam Revolutionizes Efficiency in Large Data Problems

Optimization Methods
Optimization Techniques
Extreme Learning Machines
2 views
Theme background image
Computer Science
5 years ago

DenseNet: Revolutionizing Object Recognition with Fewer Parameters and Faster Training

Computer Vision
Transfer Learning
Action Recognition
1 view