Blogs

Jigsaw based Self Supervised Learning on CIFAR-10

Initially we train EfficientNet model on 5% of the data and then test it on 50% of the test data. Then we create the jigsaw pre training dataset on 45% of the images. Then…

Fisher Information

Used to find out how much information a particular random variable has about an unknown parameter. Example: If we have a coin and we want to know if it is fair or not, we…

Change of Variables with Jacobian Matrix

The above plots demonstrate how the grid changes and figures are…

JPEG

We will explore JPEG Compression in this Blog. First we create a synthetic image of size 256*256 and we will compress it using the JPEG Algorithm.

[Paper Review] ImageNet Classification with Deep Convolutional Neural Networks (2012)

Neural Networks (2012)