Towards Lossless Encoding of Sentences
Posted on04/06/2019
A lot of work has been done in the field of image compression via machine learning, but not much attention has been... Read More
Do Neural Dialog Systems Use the Conversation History Effectively? An Empirical Study
Posted on04/06/2019
Neural generative models have been become increasingly popular when building conversational agents. They offer flexibility, can be easily adapted to new domains,... Read More
A Cross-Domain Transferable Neural Coherence Model
Posted on28/05/2019
Coherence is an important aspect of text quality and is crucial for ensuring its readability. One important limitation of existing coherence models... Read More
State-Reification Networks: Improving Generalization by Modeling the Distribution of Hidden Representations
Posted on26/05/2019
Machine learning promises methods that generalize well from finite labeled data. However, the brittleness of existing neural net approaches is revealed by... Read More
On Variational Bounds of Mutual Information
Posted on16/05/2019
Estimating and optimizing Mutual Information (MI) is core to many problems in machine learning; however, bounding MI in high dimensions is challenging.... Read More
GMNN: Graph Markov Neural Networks
Posted on15/05/2019
This paper studies semi-supervised object classification in relational data, which is a fundamental problem in relational data modeling. The problem has been... Read More
Hierarchical Importance Weighted Autoencoders
Posted on13/05/2019
Importance weighted variational inference (Burda et al., 2015) uses multiple i.i.d. samples to have a tighter variational lower bound. We believe a... Read More
Variational State Encoding as Intrinsic Motivation in Reinforcement Learning
Posted on06/05/2019
Discovering efficient exploration strategies is a central challenge in reinforcement learning (RL), especially in the context of sparse rewards environments. We postulate... Read More
Active Domain Randomization
Posted on09/04/2019
Domain randomization is a popular technique for improving domain transfer, often used in a zero-shot setting when the target domain is unknown... Read More
Speech Model Pre-training for End-to-End Spoken Language Understanding
Posted on07/04/2019
Whereas conventional spoken language understanding (SLU) systems map speech to text, and then text to intent, end-to-end SLU systems map speech directly... Read More