| Taha Ceritli and Amos Storkey |
| Tue 19 Mar 2019, 11:00 - 12:00 |
| IF 4.31/4.33 |
If you have a question about this talk, please contact: Gareth Beedham (gbeedham)
Taha Ceritli
ptype: Probabilistic Type Inference
Type inference refers to the task of inferring the data type of a given column of data. Current approaches often fail when data contains missing data and anomalies, which are found commonly in real-world data sets. In this paper, we propose ptype, a probabilistic robust type inference method that allows us to detect such entries, and infer data types. We further show that the proposed method outperforms the existing methods.
Amos Storkey
Exploration by random network distillation: efficient proxies to Bayesian predictive variances
I will motivate the idea of random network distillation as a proxy for predictive uncertainty that is well suited for neural networks. I will then very briefly outline how we used that in a reinforcement learning setting for curiosity driven learning to achieve state of the art results on Montezumas Revenge.
Joint work with Yuri Burda, Harri Edwards and Oleg Klimov.