Gather CNNs Basic Knowledge

[ghost]

Read and understand CNNs.

Questions & Problems:

[Simsso]

Tried to visualize the problem of adversarial attacks (for a better understanding). The input space is so high-dimensional that there are many data points which the network misclassifies. The reason for that is the networks complexity: It correctly classifies most of the data points, however, in between it does weird things.

Here is an example in 2D. In 10000D this gets even wilder.

[Simsso]

The Stanford lecture which Florian has mentioned today: CS231N

Computer Vision has become ubiquitous in our society, with applications in search, image understanding, apps, mapping, medicine, drones, and self-driving cars. Core to many of these applications are visual recognition tasks such as image classification, localization and detection. Recent developments in neural network (aka “deep learning”) approaches have greatly advanced the performance of these state-of-the-art visual recognition systems. This course is a deep dive into details of the deep learning architectures with a focus on learning end-to-end models for these tasks, particularly image classification. During the 10-week course, students will learn to implement, train and debug their own neural networks and gain a detailed understanding of cutting-edge research in computer vision. The final assignment will involve training a multi-million parameter convolutional neural network and applying it on the largest image classification dataset (ImageNet). We will focus on teaching how to set up the problem of image recognition, the learning algorithms (e.g. backpropagation), practical engineering tricks for training and fine-tuning the networks and guide the students through hands-on assignments and a final course project. Much of the background and materials of this course will be drawn from the ImageNet Challenge.

[ghost]

Thanks again Florian for mentioning that!

I will start it right after I finished the chapter with CNNs.

[Simsso]

Not exactly basic CNN knowledge but still relevant: Lecture 16 | Adversarial Examples and Adversarial Training (by guest lecturer Ian Goodfellow).

[ghost]

Reading completed. 📖