 Strong mixed-integer programming formulations for trained neural
networksMathematical programming (Math. Program.), 2018 |
 Multi-scale Alignment and Contextual History for Attention Mechanism in
Sequence-to-sequence ModelSpoken Language Technology Workshop (SLT), 2018 |
 |
 Approximability of Discriminators Implies Diversity in GANsInternational Conference on Learning Representations (ICLR), 2018 |
 |
