DeepfakeCapsuleGAN
Published:
Using GANS to generate images that are then used for deep fakes. Using Capsule networks instead of regular CNNs
Posts by Tags
Capsule Networks
Chatbot
Improving Open Domain Dialogue-Systems (Chatbots)
Published:
Built a seq2seq neural conversational model in PyTorch using attention with intention and a diversity promoting objective function to prevent irrelevant generic outputs’
Computer Vision
Green Screen Image Composition Transfer
Published:
Using StableDiffusion for text2img background generation and MODNET/U2net model to seamlessly superimpose the isolated foreground using Alpha-matting and apply Nvidia’s FastPhotoStyle for an image stylization and smoothening
DeepfakeCapsuleGAN
Published:
Using GANS to generate images that are then used for deep fakes. Using Capsule networks instead of regular CNNs
DDQN
DSR Reinforcement Learning to navigate a Labyrinth
Published:
Generalized Successor Representations from Neuroscience within an end-to-end deep reinforcement learning framework, comparing its efficacy to DQN on two diverse environments (Mazebase and DOOM) given raw pixel observations.’
DeepFake
DeepfakeCapsuleGAN
Published:
Using GANS to generate images that are then used for deep fakes. Using Capsule networks instead of regular CNNs
Dialogue-system
Improving Open Domain Dialogue-Systems (Chatbots)
Published:
Built a seq2seq neural conversational model in PyTorch using attention with intention and a diversity promoting objective function to prevent irrelevant generic outputs’
Embeddings
Minimally-Constrained Multilingual Embeddings via Artificial Code-Switching
Published:
Created language agnostic word embeddings via artificial code-switching to share structure across languages for any NLP task when you have less labeled data.
GANS
DeepfakeCapsuleGAN
Published:
Using GANS to generate images that are then used for deep fakes. Using Capsule networks instead of regular CNNs
Generative models
DeepfakeCapsuleGAN
Published:
Using GANS to generate images that are then used for deep fakes. Using Capsule networks instead of regular CNNs
Image Stylization
Green Screen Image Composition Transfer
Published:
Using StableDiffusion for text2img background generation and MODNET/U2net model to seamlessly superimpose the isolated foreground using Alpha-matting and apply Nvidia’s FastPhotoStyle for an image stylization and smoothening
Multi-Lingual
Minimally-Constrained Multilingual Embeddings via Artificial Code-Switching
Published:
Created language agnostic word embeddings via artificial code-switching to share structure across languages for any NLP task when you have less labeled data.
NLP
Improving Open Domain Dialogue-Systems (Chatbots)
Published:
Built a seq2seq neural conversational model in PyTorch using attention with intention and a diversity promoting objective function to prevent irrelevant generic outputs’
Minimally-Constrained Multilingual Embeddings via Artificial Code-Switching
Published:
Created language agnostic word embeddings via artificial code-switching to share structure across languages for any NLP task when you have less labeled data.
RL
DSR Reinforcement Learning to navigate a Labyrinth
Published:
Generalized Successor Representations from Neuroscience within an end-to-end deep reinforcement learning framework, comparing its efficacy to DQN on two diverse environments (Mazebase and DOOM) given raw pixel observations.’
Reinforcement Learning
DSR Reinforcement Learning to navigate a Labyrinth
Published:
Generalized Successor Representations from Neuroscience within an end-to-end deep reinforcement learning framework, comparing its efficacy to DQN on two diverse environments (Mazebase and DOOM) given raw pixel observations.’
Seq2Seq
Improving Open Domain Dialogue-Systems (Chatbots)
Published:
Built a seq2seq neural conversational model in PyTorch using attention with intention and a diversity promoting objective function to prevent irrelevant generic outputs’
StableDiffusion
Green Screen Image Composition Transfer
Published:
Using StableDiffusion for text2img background generation and MODNET/U2net model to seamlessly superimpose the isolated foreground using Alpha-matting and apply Nvidia’s FastPhotoStyle for an image stylization and smoothening