Alert button
Picture for Andrew K. Lampinen

Andrew K. Lampinen

Alert button

Zipfian environments for Reinforcement Learning

Mar 15, 2022
Stephanie C. Y. Chan, Andrew K. Lampinen, Pierre H. Richemond, Felix Hill

Figure 1 for Zipfian environments for Reinforcement Learning
Figure 2 for Zipfian environments for Reinforcement Learning
Figure 3 for Zipfian environments for Reinforcement Learning
Figure 4 for Zipfian environments for Reinforcement Learning
Viaarxiv icon

Tell me why! -- Explanations support learning of relational and causal structure

Dec 08, 2021
Andrew K. Lampinen, Nicholas A. Roy, Ishita Dasgupta, Stephanie C. Y. Chan, Allison C. Tam, James L. McClelland, Chen Yan, Adam Santoro, Neil C. Rabinowitz, Jane X. Wang, Felix Hill

Figure 1 for Tell me why! -- Explanations support learning of relational and causal structure
Figure 2 for Tell me why! -- Explanations support learning of relational and causal structure
Figure 3 for Tell me why! -- Explanations support learning of relational and causal structure
Figure 4 for Tell me why! -- Explanations support learning of relational and causal structure
Viaarxiv icon

What shapes feature representations? Exploring datasets, architectures, and training

Jun 22, 2020
Katherine L. Hermann, Andrew K. Lampinen

Figure 1 for What shapes feature representations? Exploring datasets, architectures, and training
Figure 2 for What shapes feature representations? Exploring datasets, architectures, and training
Figure 3 for What shapes feature representations? Exploring datasets, architectures, and training
Figure 4 for What shapes feature representations? Exploring datasets, architectures, and training
Viaarxiv icon

Transforming task representations to allow deep learning models to perform novel tasks

May 08, 2020
Andrew K. Lampinen, James L. McClelland

Figure 1 for Transforming task representations to allow deep learning models to perform novel tasks
Figure 2 for Transforming task representations to allow deep learning models to perform novel tasks
Figure 3 for Transforming task representations to allow deep learning models to perform novel tasks
Figure 4 for Transforming task representations to allow deep learning models to perform novel tasks
Viaarxiv icon

Automated curricula through setter-solver interactions

Sep 27, 2019
Sebastien Racaniere, Andrew K. Lampinen, Adam Santoro, David P. Reichert, Vlad Firoiu, Timothy P. Lillicrap

Figure 1 for Automated curricula through setter-solver interactions
Figure 2 for Automated curricula through setter-solver interactions
Figure 3 for Automated curricula through setter-solver interactions
Figure 4 for Automated curricula through setter-solver interactions
Viaarxiv icon

Embedded Meta-Learning: Toward more flexible deep-learning models

May 23, 2019
Andrew K. Lampinen, James L. McClelland

Figure 1 for Embedded Meta-Learning: Toward more flexible deep-learning models
Figure 2 for Embedded Meta-Learning: Toward more flexible deep-learning models
Figure 3 for Embedded Meta-Learning: Toward more flexible deep-learning models
Figure 4 for Embedded Meta-Learning: Toward more flexible deep-learning models
Viaarxiv icon

An analytic theory of generalization dynamics and transfer learning in deep linear networks

Sep 27, 2018
Andrew K. Lampinen, Surya Ganguli

Figure 1 for An analytic theory of generalization dynamics and transfer learning in deep linear networks
Figure 2 for An analytic theory of generalization dynamics and transfer learning in deep linear networks
Figure 3 for An analytic theory of generalization dynamics and transfer learning in deep linear networks
Figure 4 for An analytic theory of generalization dynamics and transfer learning in deep linear networks
Viaarxiv icon

One-shot and few-shot learning of word embeddings

Jan 02, 2018
Andrew K. Lampinen, James L. McClelland

Figure 1 for One-shot and few-shot learning of word embeddings
Figure 2 for One-shot and few-shot learning of word embeddings
Figure 3 for One-shot and few-shot learning of word embeddings
Figure 4 for One-shot and few-shot learning of word embeddings
Viaarxiv icon