Emti's photo

Emtiyaz Khan

Team Leader
Approximate Bayesian Inference (ABI) team
Center for Advanced Intelligence Project (AIP)
RIKEN, Tokyo
emtiyaz [at] gmail.com
emtiyaz.khan [at] riken.jp

I no longer update this webpage.
Please visit my new page here.

Research Publications Teaching Blog

Open positions in my team

I have several post-doc, research assistant, and intern positions. Please email me if you are interested. You might also want to see this advert for more details.

New talk at the PGM workshop 2017 in ISM

(Happening on Feb. 22, 2017) I will give a talk at the Probabilistic Graphical Model Workshop 2017 to be held in the Institute of Statistical Mathematics, Tokyo.

Poster presentation at the Winter-Festa

(Dec. 22, 2017) I presented a poster at the Winter-Festa (held in Tokyo) on converting variational inference (VI) in complex models to VI in simple conjugate-models. See my "hand-made" poster.

New paper on Deep Exponential Family (DEF)

(Dec. 2017) We presented a paper at Bayesian Deep Learning workshop in NIPS 2016 for inference in DEFs using stochastic conjugate computations similar to variational message passing.

New job! New team!

(Oct. 2016) I started working as a Team Leader at RIKEN's newly established Center for Advanced Intelligence Project (AIP). I lead the Approximate Bayesian Inference (ABI) team.

New paper on Voting Data Analysis

(June 2016) A new paper at DSAA, 2016.

New paper on Stochastic Variational Inference

(June 2016) A new paper at UAI, 2016.

Teaching Award!

(Dec. 2015) I got a teaching award at IC department in EPFL for teaching the Machine-Learning course!


The main goal of my team is to understand the principles of learning from data, and to use that understanding to develop algorithms that can learn like living beings. Currently, our focus is to understand the role of uncertainty in learning and to develop fast algorithms for uncertainty estimation.

We are working on the following research projects:

  • Variational inference for large and complex models.
  • Stochastic algorithms for Bayesian deep learning.
  • Scalable inference for Gaussian process models.
  • Automating Data Science.

We are working on the following application projects:

  • Machine learning for the design of high-performance buildings.
  • UAVs doing Bayesian optimization to track humans.
  • Context-aware and automatic permissions for mobile devices.
  • Online collaborative predictions of vote results.