Published Jan 15, 2015
Easily Fooling Deep Neural Networks
Explore the vulnerabilities and reliability concerns of deep neural networks as Anh Nguyen discusses how these systems can be easily fooled by images, despite their advancements in fields like medical imaging and 3D analysis. Gain insights into the parallels between AI's optical illusions and human perception, emphasizing the need for more robust training methods to ensure AI's reliability in sensitive applications.

Topics covered
Popular Clips
Episode Highlights
Related Episodes


Fooling Computer Vision
Answers 383 questions

Visual Illusions Deceiving Neural Networks
Answers 383 questions

Understanding Neural Networks
Answers 383 questions

The Complexity of Learning Neural Networks
Answers 383 questions
[MINI] Convolutional Neural Networks
Answers 383 questions
Deep Fakes
Answers 383 questions

Computer Vision is Not Perfect
Answers 383 questions

Neuroscience from a Data Scientist's Perspective
Answers 383 questions

Machine Learning Done Wrong
Answers 383 questions

Modeling Fake News
Answers 383 questions

Reproducing Deep Learning Models
Answers 383 questions

Human Detection of Fake News
Answers 383 questions

[MINI] Dropout
Answers 383 questions
Artificial Intelligence, a Podcast Approach
Answers 383 questions
[MINI] Natural Language Processing
Answers 383 questions
