Picture for Clare Lyle

Clare Lyle

Generalization Through the Lens of Learning Dynamics

Dec 11, 2022
Clare Lyle

* PhD Thesis

Access Paper or Ask Questions

Understanding and Preventing Capacity Loss in Reinforcement Learning

Apr 20, 2022
Clare Lyle, Mark Rowland, Will Dabney

Figure 1 for Understanding and Preventing Capacity Loss in Reinforcement Learning

Figure 2 for Understanding and Preventing Capacity Loss in Reinforcement Learning

Figure 3 for Understanding and Preventing Capacity Loss in Reinforcement Learning

Figure 4 for Understanding and Preventing Capacity Loss in Reinforcement Learning

* Presented at ICLR 2022

Access Paper or Ask Questions

DARTS without a Validation Set: Optimizing the Marginal Likelihood

Dec 24, 2021
Miroslav Fil, Binxin Ru, Clare Lyle, Yarin Gal

Figure 1 for DARTS without a Validation Set: Optimizing the Marginal Likelihood

Figure 2 for DARTS without a Validation Set: Optimizing the Marginal Likelihood

Figure 3 for DARTS without a Validation Set: Optimizing the Marginal Likelihood

Figure 4 for DARTS without a Validation Set: Optimizing the Marginal Likelihood

* Presented at the 5th Workshop on Meta-Learning at NeurIPS 2021

Access Paper or Ask Questions

Self-Attention Between Datapoints: Going Beyond Individual Input-Output Pairs in Deep Learning

Jun 04, 2021
Jannik Kossen, Neil Band, Clare Lyle, Aidan N. Gomez, Tom Rainforth, Yarin Gal

Figure 1 for Self-Attention Between Datapoints: Going Beyond Individual Input-Output Pairs in Deep Learning

Figure 2 for Self-Attention Between Datapoints: Going Beyond Individual Input-Output Pairs in Deep Learning

Figure 3 for Self-Attention Between Datapoints: Going Beyond Individual Input-Output Pairs in Deep Learning

Figure 4 for Self-Attention Between Datapoints: Going Beyond Individual Input-Output Pairs in Deep Learning

* First two authors contributed equally

Access Paper or Ask Questions

Provable Guarantees on the Robustness of Decision Rules to Causal Interventions

May 19, 2021
Benjie Wang, Clare Lyle, Marta Kwiatkowska

Figure 1 for Provable Guarantees on the Robustness of Decision Rules to Causal Interventions

Figure 2 for Provable Guarantees on the Robustness of Decision Rules to Causal Interventions

Figure 3 for Provable Guarantees on the Robustness of Decision Rules to Causal Interventions

Figure 4 for Provable Guarantees on the Robustness of Decision Rules to Causal Interventions

* 21 pages (8+13 Appendix). To be published in IJCAI 2021

Access Paper or Ask Questions

On The Effect of Auxiliary Tasks on Representation Dynamics

Feb 25, 2021
Clare Lyle, Mark Rowland, Georg Ostrovski, Will Dabney

Figure 1 for On The Effect of Auxiliary Tasks on Representation Dynamics

Figure 2 for On The Effect of Auxiliary Tasks on Representation Dynamics

Figure 3 for On The Effect of Auxiliary Tasks on Representation Dynamics

Figure 4 for On The Effect of Auxiliary Tasks on Representation Dynamics

* AISTATS 2021

Access Paper or Ask Questions