Training energy-based models (EBMs) on discrete spaces is challenging be...
Energy-based models are a simple yet powerful class of probabilistic mod...
A key challenge in video question answering is how to realize the cross-...
Learning set functions becomes increasingly more important in many
appli...
Recent years have witnessed the dramatic growth of paper volumes with pl...
Existing unsupervised document hashing methods are mostly established on...
With the need of fast retrieval speed and small memory footprint, docume...
Many unsupervised hashing methods are implicitly established on the idea...
For task-oriented dialog systems, training a Reinforcement Learning (RL)...