Posts

Paper Summary: Uniform Convergence May Be Unable to Explain Generalization in Deep Learning (NeurIPS 19)

This paper proposes a novel technique for learning node representations and at the same time perform community detection task for the graphical data by creating a generative model using the variational inference concepts.

Deep Learning Winter Reading Series: Part 1

Winters bring a lot of free time for you to catch up with amazing stuff. The Vision and Language Group will be posting learning resources on a weekly basis, comprising both interesting reads and talks on future research areas.

Paper Summary: This Looks Like That: Deep Learning for Interpretable Image Recognition (NeurIPS 19)

This paper proposes a novel idea for interpretable deep learning , it basically figures out some protopyical parts of images by itself , and then uses these prototypes to make classification , hence making the classification process interpretable.

Paper Summary: vGraph: A Generative Model for Joint Community Detection and Node Representational Learning (NeurIPS 19)

This paper proposes a novel technique for learning node representations and at the same time perform community detection task for the graphical data by creating a generative model using the variational inference concepts.

Essential Deep Learning Topics for Interviews

This post contains a list of topics which we feel that one should be comfortable with before appearing for a DL interview. This list is by no means exhaustive (as the field is very wide and ever growing).

Paper Summary: SinGAN Learning a Generative Model from a Single Natural Image (ICCV 19 Best Paper)

This paper proposes a novel GAN training technique to obtain a generative model that can be learned using a single image. Unlike some of the previous works that used single image for training for a single task, the model can be used for unconditional generative modelling, not limited to texture images.