Talks
All the talks so far.
Input Combination Strategies for Multi-Source Transformer Decoder
Nalin KumarPublished: at 03:00 PMIn this talk, we will discuss the different input combination strategies for the encoder-decoder attention in the Transformer architecture.
Optimizing Transformer for Low-Resource Neural Machine Translation
Deepak KumarPublished: at 03:00 PMIn this talk, we will look at the effectiveness of the Transformer model under low-resource conditions and how the translation quality can be improved by optimizing the hyper-parameters.
MIME - MIMicking Emotions for Empathetic Response Generation
Deepak KumarPublished: at 12:00 PMIn this talk, we will discuss the importance of polarity-based emotion clusters and emotional mimicry in empathetic response generation.
The Math of Differentiable Rendering
Annada Prasad BeheraPublished: at 12:00 PMIn this talk, we will talk about the rendering and anti-aliasing equations and their approximations in real-world rendering. We will also discuss how to preserve differentiability in rendering using Feynman's differentiation under the integral sign and the Reynolds Transport Theorem.
Data Augmentation for Neural Machine Translation (NMT)
Nalin KumarPublished: at 07:00 PMIn this talk, we will discuss various data augmentation techniques used for Neural Machine Translation.
Synthesis of Novel Views by Estimating Radiance Fields
Annada Prasad BeheraPublished: at 07:00 PMIn this talk, we will discuss different aspects of neural rendering.