582 citations · 739 across the 8 of their papers we have counts for
Showing 2015 · cs.LGShow all
2 papers · 2 filters
cs.LG2015★ 87 cited
Scalable Gradient-Based Tuning of Continuous Regularization Hyperparameters
Jelena Luketina, Mathias Berglund, Klaus Greff +1
Hyperparameter selection generally relies on running multiple full training trials, with selection based on validation set performance. We propose a gradient-based approach for loc…
cs.LG2015★ 18 cited
Bidirectional Recurrent Neural Networks as Generative Models - Reconstructing Gaps in Time Series
Mathias Berglund, Tapani Raiko, Mikko Honkala +3
Bidirectional recurrent neural networks (RNN) are trained to predict both in the positive and negative time directions simultaneously. They have not been used commonly in unsupervi…