Alert button
Picture for Andrea Gesmundo

Andrea Gesmundo

Alert button

Composable Function-preserving Expansions for Transformer Architectures

Aug 11, 2023
Andrea Gesmundo, Kaitlin Maile

Viaarxiv icon

Multipath agents for modular multitask ML systems

Feb 06, 2023
Andrea Gesmundo

Figure 1 for Multipath agents for modular multitask ML systems
Figure 2 for Multipath agents for modular multitask ML systems
Figure 3 for Multipath agents for modular multitask ML systems
Figure 4 for Multipath agents for modular multitask ML systems
Viaarxiv icon

A Multi-Agent Framework for the Asynchronous and Collaborative Extension of Multitask ML Systems

Sep 29, 2022
Andrea Gesmundo

Figure 1 for A Multi-Agent Framework for the Asynchronous and Collaborative Extension of Multitask ML Systems
Figure 2 for A Multi-Agent Framework for the Asynchronous and Collaborative Extension of Multitask ML Systems
Viaarxiv icon

A Continual Development Methodology for Large-scale Multitask Dynamic ML Systems

Sep 15, 2022
Andrea Gesmundo

Figure 1 for A Continual Development Methodology for Large-scale Multitask Dynamic ML Systems
Figure 2 for A Continual Development Methodology for Large-scale Multitask Dynamic ML Systems
Figure 3 for A Continual Development Methodology for Large-scale Multitask Dynamic ML Systems
Figure 4 for A Continual Development Methodology for Large-scale Multitask Dynamic ML Systems
Viaarxiv icon

An Evolutionary Approach to Dynamic Introduction of Tasks in Large-scale Multitask Learning Systems

Jun 05, 2022
Andrea Gesmundo, Jeff Dean

Figure 1 for An Evolutionary Approach to Dynamic Introduction of Tasks in Large-scale Multitask Learning Systems
Figure 2 for An Evolutionary Approach to Dynamic Introduction of Tasks in Large-scale Multitask Learning Systems
Figure 3 for An Evolutionary Approach to Dynamic Introduction of Tasks in Large-scale Multitask Learning Systems
Figure 4 for An Evolutionary Approach to Dynamic Introduction of Tasks in Large-scale Multitask Learning Systems
Viaarxiv icon

muNet: Evolving Pretrained Deep Neural Networks into Scalable Auto-tuning Multitask Systems

May 25, 2022
Andrea Gesmundo, Jeff Dean

Figure 1 for muNet: Evolving Pretrained Deep Neural Networks into Scalable Auto-tuning Multitask Systems
Figure 2 for muNet: Evolving Pretrained Deep Neural Networks into Scalable Auto-tuning Multitask Systems
Figure 3 for muNet: Evolving Pretrained Deep Neural Networks into Scalable Auto-tuning Multitask Systems
Figure 4 for muNet: Evolving Pretrained Deep Neural Networks into Scalable Auto-tuning Multitask Systems
Viaarxiv icon

Scaling Up Models and Data with $\texttt{t5x}$ and $\texttt{seqio}$

Mar 31, 2022
Adam Roberts, Hyung Won Chung, Anselm Levskaya, Gaurav Mishra, James Bradbury, Daniel Andor, Sharan Narang, Brian Lester, Colin Gaffney, Afroz Mohiuddin, Curtis Hawthorne, Aitor Lewkowycz, Alex Salcianu, Marc van Zee, Jacob Austin, Sebastian Goodman, Livio Baldini Soares, Haitang Hu, Sasha Tsvyashchenko, Aakanksha Chowdhery, Jasmijn Bastings, Jannis Bulian, Xavier Garcia, Jianmo Ni, Andrew Chen, Kathleen Kenealy, Jonathan H. Clark, Stephan Lee, Dan Garrette, James Lee-Thorp, Colin Raffel, Noam Shazeer, Marvin Ritter, Maarten Bosma, Alexandre Passos, Jeremy Maitin-Shepard, Noah Fiedel, Mark Omernick, Brennan Saeta, Ryan Sepassi, Alexander Spiridonov, Joshua Newlan, Andrea Gesmundo

Figure 1 for Scaling Up Models and Data with $\texttt{t5x}$ and $\texttt{seqio}$
Figure 2 for Scaling Up Models and Data with $\texttt{t5x}$ and $\texttt{seqio}$
Viaarxiv icon

Routing Networks with Co-training for Continual Learning

Sep 09, 2020
Mark Collier, Efi Kokiopoulou, Andrea Gesmundo, Jesse Berent

Figure 1 for Routing Networks with Co-training for Continual Learning
Figure 2 for Routing Networks with Co-training for Continual Learning
Figure 3 for Routing Networks with Co-training for Continual Learning
Figure 4 for Routing Networks with Co-training for Continual Learning
Viaarxiv icon