Alert button
Picture for Debadeepta Dey

Debadeepta Dey

Alert button

Neural Architecture Search: Insights from 1000 Papers

Jan 25, 2023
Colin White, Mahmoud Safari, Rhea Sukthanker, Binxin Ru, Thomas Elsken, Arber Zela, Debadeepta Dey, Frank Hutter

Figure 1 for Neural Architecture Search: Insights from 1000 Papers
Figure 2 for Neural Architecture Search: Insights from 1000 Papers
Figure 3 for Neural Architecture Search: Insights from 1000 Papers
Figure 4 for Neural Architecture Search: Insights from 1000 Papers
Viaarxiv icon

What Makes Convolutional Models Great on Long Sequence Modeling?

Oct 17, 2022
Yuhong Li, Tianle Cai, Yi Zhang, Deming Chen, Debadeepta Dey

Figure 1 for What Makes Convolutional Models Great on Long Sequence Modeling?
Figure 2 for What Makes Convolutional Models Great on Long Sequence Modeling?
Figure 3 for What Makes Convolutional Models Great on Long Sequence Modeling?
Figure 4 for What Makes Convolutional Models Great on Long Sequence Modeling?
Viaarxiv icon

Small Character Models Match Large Word Models for Autocomplete Under Memory Constraints

Oct 06, 2022
Ganesh Jawahar, Subhabrata Mukherjee, Debadeepta Dey, Muhammad Abdul-Mageed, Laks V. S. Lakshmanan, Caio Cesar Teodoro Mendes, Gustavo Henrique de Rosa, Shital Shah

Figure 1 for Small Character Models Match Large Word Models for Autocomplete Under Memory Constraints
Figure 2 for Small Character Models Match Large Word Models for Autocomplete Under Memory Constraints
Figure 3 for Small Character Models Match Large Word Models for Autocomplete Under Memory Constraints
Figure 4 for Small Character Models Match Large Word Models for Autocomplete Under Memory Constraints
Viaarxiv icon

One Network Doesn't Rule Them All: Moving Beyond Handcrafted Architectures in Self-Supervised Learning

Mar 15, 2022
Sharath Girish, Debadeepta Dey, Neel Joshi, Vibhav Vineet, Shital Shah, Caio Cesar Teodoro Mendes, Abhinav Shrivastava, Yale Song

Figure 1 for One Network Doesn't Rule Them All: Moving Beyond Handcrafted Architectures in Self-Supervised Learning
Figure 2 for One Network Doesn't Rule Them All: Moving Beyond Handcrafted Architectures in Self-Supervised Learning
Figure 3 for One Network Doesn't Rule Them All: Moving Beyond Handcrafted Architectures in Self-Supervised Learning
Figure 4 for One Network Doesn't Rule Them All: Moving Beyond Handcrafted Architectures in Self-Supervised Learning
Viaarxiv icon

LiteTransformerSearch: Training-free On-device Search for Efficient Autoregressive Language Models

Mar 04, 2022
Mojan Javaheripi, Shital Shah, Subhabrata Mukherjee, Tomasz L. Religa, Caio C. T. Mendes, Gustavo H. de Rosa, Sebastien Bubeck, Farinaz Koushanfar, Debadeepta Dey

Figure 1 for LiteTransformerSearch: Training-free On-device Search for Efficient Autoregressive Language Models
Figure 2 for LiteTransformerSearch: Training-free On-device Search for Efficient Autoregressive Language Models
Figure 3 for LiteTransformerSearch: Training-free On-device Search for Efficient Autoregressive Language Models
Figure 4 for LiteTransformerSearch: Training-free On-device Search for Efficient Autoregressive Language Models
Viaarxiv icon

AutoDistil: Few-shot Task-agnostic Neural Architecture Search for Distilling Large Language Models

Jan 29, 2022
Dongkuan Xu, Subhabrata Mukherjee, Xiaodong Liu, Debadeepta Dey, Wenhui Wang, Xiang Zhang, Ahmed Hassan Awadallah, Jianfeng Gao

Figure 1 for AutoDistil: Few-shot Task-agnostic Neural Architecture Search for Distilling Large Language Models
Figure 2 for AutoDistil: Few-shot Task-agnostic Neural Architecture Search for Distilling Large Language Models
Figure 3 for AutoDistil: Few-shot Task-agnostic Neural Architecture Search for Distilling Large Language Models
Figure 4 for AutoDistil: Few-shot Task-agnostic Neural Architecture Search for Distilling Large Language Models
Viaarxiv icon

FEAR: A Simple Lightweight Method to Rank Architectures

Jun 07, 2021
Debadeepta Dey, Shital Shah, Sebastien Bubeck

Figure 1 for FEAR: A Simple Lightweight Method to Rank Architectures
Figure 2 for FEAR: A Simple Lightweight Method to Rank Architectures
Figure 3 for FEAR: A Simple Lightweight Method to Rank Architectures
Figure 4 for FEAR: A Simple Lightweight Method to Rank Architectures
Viaarxiv icon

Reparameterized Variational Divergence Minimization for Stable Imitation

Jun 18, 2020
Dilip Arumugam, Debadeepta Dey, Alekh Agarwal, Asli Celikyilmaz, Elnaz Nouri, Bill Dolan

Figure 1 for Reparameterized Variational Divergence Minimization for Stable Imitation
Figure 2 for Reparameterized Variational Divergence Minimization for Stable Imitation
Figure 3 for Reparameterized Variational Divergence Minimization for Stable Imitation
Figure 4 for Reparameterized Variational Divergence Minimization for Stable Imitation
Viaarxiv icon

A Recipe for Creating Multimodal Aligned Datasets for Sequential Tasks

May 19, 2020
Angela S. Lin, Sudha Rao, Asli Celikyilmaz, Elnaz Nouri, Chris Brockett, Debadeepta Dey, Bill Dolan

Figure 1 for A Recipe for Creating Multimodal Aligned Datasets for Sequential Tasks
Figure 2 for A Recipe for Creating Multimodal Aligned Datasets for Sequential Tasks
Figure 3 for A Recipe for Creating Multimodal Aligned Datasets for Sequential Tasks
Figure 4 for A Recipe for Creating Multimodal Aligned Datasets for Sequential Tasks
Viaarxiv icon

Efficient Forward Architecture Search

May 31, 2019
Hanzhang Hu, John Langford, Rich Caruana, Saurajit Mukherjee, Eric Horvitz, Debadeepta Dey

Figure 1 for Efficient Forward Architecture Search
Figure 2 for Efficient Forward Architecture Search
Figure 3 for Efficient Forward Architecture Search
Figure 4 for Efficient Forward Architecture Search
Viaarxiv icon