Update

news

An rational approximation for graph embeddings

Research

work in progress

Ceative music generation

The goal is to produce new music at human musician’s level

Geometric Deep Learning

Ultilize deep learning technique to model graph/networks

Multimodal Learning

Extract shared features by multiple representations.

Understanding DNN

understanding deep learning by math and physics

Selected Publications

A rational neural networks for improving graph node-level embeddings
ICDM, 2018

Publications

More Publications

. StreetNet: Preference Learning with Convolutional Neural Network on Urban Crime Perception. SIGSPATIAL, 2018.

. DIMPL: A Distributed In-Memory Drone Flight Path Builder System. Big Data*, 2018.

PDF

. Distributed Self-Paced Learning in Alternating Direction Method of Multipliers. IJCAI, 2018.

Preprint

. Learning to Fuse Music Genres with Generative Adversarial Dual Learning. ICDM, 2017.

Preprint PDF Code Slides

. TRACES: Generating Twitter Stories via Shared Subspace and Temporal Smoothness. Big Data, 2017.

PDF

. Trendi:Tracking Stories in News and Microblogs via Emerging, Evolving and Fading Topics. Big Data, 2017.

PDF

. Virtual Metering: An Efficient Water Disaggregation Algorithm via Non-Intrusive Load Monitoring. TIST, 2017.

PDF

. Multimodal Storytelling via Generative Adversarial Imitation Learning. IJCAI, 2017.

Preprint PDF

. Storytelling in heterogeneous Twitter entity network based on hierarchical cluster routing. Big Data, 2016.

PDF

. DIFPLL: Distributed drone flight path builder system. (Best student paper). GISTAM, 2015.

PDF