Poster
Sequence Tutor: Conservative fine-tuning of sequence generation models with KL-control
Natasha Jaques · Shixiang Gu · Dzmitry Bahdanau · Jose Hernandez-Lobato · Richard E Turner · Douglas Eck

Tue Aug 8th 06:30 -- 10:00 PM @ Gallery #77

This paper proposes a general method for improving the structure and quality of sequences generated by a recurrent neural network (RNN), while maintaining information originally learned from data, as well as sample diversity. An RNN is first pre-trained on data using maximum likelihood estimation (MLE), and the probability distribution over the next token in the sequence learned by this model is treated as a prior policy. Another RNN is then trained using reinforcement learning (RL) to generate higher-quality outputs that account for domain-specific incentives while retaining proximity to the prior policy of the MLE RNN. To formalize this objective, we derive novel off-policy RL methods for RNNs from KL-control. The effectiveness of the approach is demonstrated on two applications; 1) generating novel musical melodies, and 2) computational molecular generation. For both problems, we show that the proposed method improves the desired properties and structure of the generated sequences, while maintaining information learned from data.

Author Information

Natasha Jaques (Massachusetts Institute of Technology)
Shixiang Gu (Cambridge)
Dzmitry Bahdanau (Université de Montréal)
Jose Hernandez-Lobato (University of Cambridge)
Richard E Turner (University of Cambridge)

Richard Turner holds a Lectureship (equivalent to US Assistant Professor) in Computer Vision and Machine Learning in the Computational and Biological Learning Lab, Department of Engineering, University of Cambridge, UK. He is a Fellow of Christ's College Cambridge. Previously, he held an EPSRC Postdoctoral research fellowship which he spent at both the University of Cambridge and the Laboratory for Computational Vision, NYU, USA. He has a PhD degree in Computational Neuroscience and Machine Learning from the Gatsby Computational Neuroscience Unit, UCL, UK and a M.Sci. degree in Natural Sciences (specialism Physics) from the University of Cambridge, UK. His research interests include machine learning, signal processing and developing probabilistic models of perception.

Douglas Eck (Google Brain)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors