The following are papers that caught my eye at this year’s ICML. If there are any awesome ones I missed, let me know.

In particular, Austerity in MCMC Land is an interesting result showing one a great way to make MCMC scalable.

---

Efficient Continuous-Time Markov Chain Estimation
Monir Hajiaghayi, Bonnie Kirkpatrick, Liangliang Wang, Alexandre Bouchard-Côté

Spherical Hamiltonian Monte Carlo for Constrained Target Distributions
Shiwei Lan, Bo Zhou, Babak Shahbaba

Hamiltonian Monte Carlo Without Detailed Balance
Jascha Sohl-Dickstein, Mayur Mudigonda, Michael DeWeese

Discriminative Features via Generalized Eigenvectors
Nikos Karampatziakis, Paul Mineiro

Quasi-Monte Carlo Feature Maps for Shift-Invariant Kernels
Jiyan Yang, Vikas Sindhwani, Haim Avron, Michael Mahoney

A reversible infinite HMM using normalised random measures
Konstantina Palla, David A. Knowles, Zoubin Ghahramani

Max-Margin Infinite Hidden Markov Models
Aonan Zhang, Jun Zhu, Bo Zhang

Online Bayesian Passive-Aggressive Learning
Tianlin Shi, Jun Zhu

Austerity in MCMC Land: Cutting the Metropolis-Hastings Budget
Anoop Korattikara, Yutian Chen, Max Welling