Learning to Party: Celebrating iiD

In early November, iiD  got together with the Duke Social Science Reseach Initiative and the Duke Energy Initiative  to celebrate recent achievements and give thanks for support.

We wanted to thank the Ahmadieh family for their support of programs and partnerships across Duke.

We also wanted to note the dozen Duke-authored papers accepted for presentation at the 29th annual Conference on Neural Information Processing Systems.

The accepted papers are listed below.

Policy Evaluation Using the Ω-Return
Philip Thomas; George Konidaris, Duke; Scott Niekum, UT Austin; Georgios Theocharous, Adobe

Parallelizing MCMC with Random Partition Trees (PDF)
Xiangyu Wang, Duke University; Fangjian Guo, Duke University; Katherine Heller, Duke University; David Dunson, Duke University

Discriminative Robust Transformation Learning (PDF)
Jiaji Huang, Duke University; Qiang Qiu, Duke University; Guillermo Sapiro; Robert Calderbank, Duke University

Fast Second Order Stochastic Backpropagation for Variational Inference (PDF)
Kai Fan, Duke University; Ziteng Wang, ; Jeff Beck; James Kwok, Hong Kong University of Science and Technology; Katherine Heller, Duke

Probabilistic Curve Learning: Coulomb Repulsion and the Electrostatic Gaussian Process (PDF)
Ye Wang, Duke University; David Dunson, Duke University

GP Kernels for Cross-Spectrum Analysis (PDF)
Kyle Ulrich, Duke; David Carlson; Lawrence Carin, Duke University

On the Convergence of Stochastic Gradient MCMC Algorithms with High-Order Integrators (PDF)
Changyou Chen, Duke University; Nan Ding, Google; Lawrence Carin, Duke University

On the Consistency Theory of High Dimensional Variable Screening
Xiangyu Wang, Duke University; Chenlei Leng; David Dunson, Duke University

Deep Temporal Sigmoid Belief Networks for Sequence Modeling (PDF)
Zhe Gan, Duke University; Chunyuan Li, Duke University; Ricardo Henao, Duke University; David Carlson; Lawrence Carin, Duke University

Deep Poisson Factor Modeling (PDF)
Ricardo Henao, Duke University; Zhe Gan, Duke University; James Lu, Duke University; Lawrence Carin, Duke University

Preconditioned Spectral Descent for Deep Learning
David Carlson; Edo Collins, ; Ya-Ping Hsieh, EPFL; Lawrence Carin, Duke University; Volkan Cevher, EPFL

Large-Scale Bayesian Multi-Label Learning via Positive Labels Only
Piyush Rai, Duke University; Changwei Hu; Ricardo Henao, Duke University; Lawrence Carin, Duke University