November 16, 2015 – A paper titled Rényi Divergence Minimization based Co-regularized Multiview Clustering that I worked on with Shalmali Joshi, Joydeep Ghosh, and Samni Koyejo has been accepted to the Machine Learning Journal ECML/PKDD 2015 Special Issue.
November 13, 2015 – I gave a high-level overview of machine learning problems and techniques at an Introduction to Machine Learning for Biology workshop at the ANU Research School of Biology.
November 4, 2015 – My student Avraham Ruderman has submitted his PhD dissertation for examination!
October 5, 2015 – Tim van Erven, Nishant Mehta, Peter Grünwald, Bob Williamson and I have a paper on Fast Rates in Statistical and Online Learning in the JMLR special issue dedicated to the memory of Alexey Chervonenkis.
September 5, 2015 – My paper, Convergence Analysis of Prediction Markets via Randomized Subspace Descent, with Rafael Frongillo was accepted to NIPS!
August 6, 2015 – I am chairing the research track of the 2nd International Conference on Predictive APIs and Applications which is being held in Sydney, Australia this year.
July 21, 2015 – Stephen Gould and I gave the first lecture for our new course, The Craft of Computing. Despite a couple of technical glitches I think it went pretty well and it looks like we have a keen cohort of students too.
July 16, 2015 – I gave a seminar to NICTA’s Machine Learning Group. I presented summaries of the work I presented at COLT (with Rafael Frongillo, Bob Willimson, and Nishant Mehta) on generalized mixability and some more recent work with Raf on risk networks.
July 9, 2015 – Tim van Erven, Peter Grünwald, Nishant Mehta, Bob Willimson, and I have just finished a long paper that relates various conditions for fast rates in online and statstical learning. The pre-print is available on arXiv.
July 3, 2015 – I am (in absentia) part of a panel on Remapping cultural history? Digital Humanities, Historical Bibliometrics, and the Reception of Print Culture with several other panelists, including my co-author Julieanne Lamond.
July 2, 2015 – I will be in Paris and Lille in France for COLT and ICML, respectively, over the next two weeks.
May 4, 2015 – My paper on Generalized Mixability via Entropic Duality with Rafael Frongillo, Nishant Mehta, and Bob Williamson has been accepted to COLT 2015. See you in Paris!
March 26, 2015 – I am please to be part of the Program Committee for the ICML MLOSS Workshop, to be held in Lille, France on July 10th, 2015.
March 10, 2015 – The international conference on Predcitive APIS and Apps will be held in Sydney on August 6–7th this year, right after KDD. I will be involved as the Research Chair. Please consider attending or, better yet, submitting a proposal. Hope to see you there!
February 25, 2015 – I am presenting a tutorial on prediction markets and their connections to machine learning at the Machine Learning Summer School in Sydney.
January 23, 2015 – I will be serving on the Program Committee for the UAI conference to be held in Amsterdam in July, 2015.
December 12, 2014 – Rafael Frongillo will present our paper on Randomized Subspace Descent at the NIPS 2014 Optimization Workshop.
December 12, 2014 – Amos Storkey and Jake Abernethy are graciously running our Workshop on Transactional Machine Learning and E-Commerce at NIPS this year. I’m really disappointed I will be missing the fantastic line-up of speakers. Rafael will be presenting some of our joint work on risk networks there.
December 11, 2014 – Stephen Gould and I have been awarded a Vice-Chancellor’s Teaching Enhancement Grant to help develop CodeBench, an online, interactive environment for our Craft of Computing course.
November 19, 2014 – I am visiting Jun Zhu and Jie Zhou at Tsinghua University until the 3rd of December as part of the Australia-China Young Scientist Exchange Program. I was one of 15 academics selected from Australian universities to be part of the exchange.
September 19, 2014 – I am very pleased to announce that Mindika Premachandra has submitted her PhD thesis for review. My first PhD student to do so. Congratulations Mindika!
September 13, 2014 – I helped judge the fantasic array of school students’ technology projects as part of the ACT Young ICT Explorers competition.
August 28, 2014 – I participated as a judge in the ANU College of Engineering 3 Minute Thesis competition and saw a great range of very high-quality PhD talks.
August 19, 2014 – Amos Storkey, Jake Abernethy and I have had our proposal for a NIPS Workshop on Transactional Machine Learning and E-Commerce accepted. Please check the website for details of speakers and submissions.
June 22, 2014 – I will be attending ICML in Beijing from June 22–26 where I am helping run the Workshop on Divergence Methods for Probabilistic Inference.
June 10, 2014 – I will be attending COLT in Barcelona from June 13–15th.
May 3, 2014 – Microsoft Research, New York is kindly hosting me as a visitor from May 5–30th while I work on projects related to DECRA.
February 14, 2014 – My paper on a hybrid loss for multiclass prediction with Qinfeng Shi, Tiberio Caetano, Anton van den Hengel, and Zhenhau Wang has published in PAMI (arXiv preprint).
February 6, 2014 – I will be co-organising an Workshop on Divergence Methods and Probabilistic Inference with Sanmi Koyejo, Jun Zhu, and Eric Xing at ICML 2014.
January 20, 2014 – I will be serving as a Program Chair for 2014 Uncertainty in AI Conference, to be held in Quebec in July, 2014.
January 15, 2014 – My paper on multiclass logitboost variant with Peng Sun (孙鹏) and Jie Zhou has been accepted to the Machine Learning Journal.
December 16, 2013 – I have had two papers accepted for oral presentation at MaxEnt 2013: one on Convex Foundations of Generalized MaxEnt Models with Rafael Frongillo, and the other on Conjugate Priors for Generalized MaxEnt Families with Brendan van Rooyen.
November 14, 2013 – James Montgomery and I have put code for a prototype PSI client and server on GitHub
October 7, 2013 – The PSI project was awarded an Amazon AWS in Education grant to help us set up PSI services on AWS.
October 3, 2013 – I presented a talk titled “A Convex Look at Exponential Families and Inference” to Peter Bartlett’s group at QUT.
October 3, 2013 – Two of my submissions to MaxEnt 2013 – one with Rafael Frongillo, the other with Brendan van Rooyen – have been accepted as oral presentations.
September 26, 2013 – Mindika Premachandra and I had our paper “Aggregating Predictions via Sequential Mini-Trading” accepted to ACML 2013.
May 2, 2013 – I presented a short talk on Prediction Markets, Aggregation, and Optimisation at the NICTA Machine Learning and Optimisation Workshop in Sydney.
April 22, 2013 – I will be visiting Tsinghua University, Beijing from the 7–10th of May as part of a Australia-China Science and Research Fund group mission.
April 2, 2013 – I have officially started work on my DECRA-funded project: Composing Machine Learning via Market Mechanisms.
March 27, 2013 – I am now an Action Editor for the Journal of Machine Learning Research’s Machine Learning Open Source Software (MLOSS) track.
November 5, 2012 – My grant proposal titled Composing Machine Learning via Market Mechanisms was successful! I will be receiving funding under the Discovery Early Career Researcher Award (DECRA) scheme for three years, starting in 2013.
October 11, 2012 – A proposal to demonstrate an implementaiton of the Protocols and Structures for Inference specification at NIPS 2012 was accepted (joint work with James Montgomery and CISRA)
September 4, 2012 – I have two papers accepted at NIPS: Mixability in Statistical Learning (with Tim van Erven, Peter Grünwald, and Bob Williamson) and Interpreting Prediction Markets: A Stochastic Approach (with Rafael Frongillo and Nicolás Della Penna).
June 8, 2012 – My JMLR paper — Mixability is Bayes Risk Curavture Relative to Log Loss — with Tim van Erven and Bob Williamson has been published.
May 1, 2012 – My three ICML 2012 submissions (with various co-authors) were accepted. See my publications page for more details. See you in Edinburgh!
March 8, 2012 – The video of my talk, Anatomy of a Learning Problem at the Relations Between Machine Learning Problems NIPS Workshop is up at videolectures.
January 31, 2012 – My paper with Nicolás Della Penna, Crowds & Prejudice: An Impossibility Theorem for Crowd Labelling with a Gold Standard was accepted to Collective Intelligence 2012.
December 10, 2011 – Travelling to NIPS. See you in Granada!
October 31, 2011 – Peng Sun, a PhD student at Tsinghua University, is visiting the machine learning group here for the next six months. We will be working on multiclass boosting.
August 29, 2011 – The Relations Between Machine Learning Problems workshop I am co-organising at NIPS is seeking submissions (due 30 Sept. 2011).
June 8, 2011 – The ML Discuss site is up for ICML 2011.
May 25, 2011 – I was interviewed for the Scientist in Schools newsletter about my involvement in the Mathematicians in Schools program.
May 10, 2011 – I have set up a page for the Reading Group I started in April. We are currently reading The Minimum Description Length Principle.
May 3, 2011 – My COLT 2011 submission with Tim and Bob was accepted! Mixability is Bayes Risk Curvature Relative to Log Loss.
April 6, 2011 – Bob and my paper, Information, Divergence and Risk for Binary Experiments has (finally!) been published in JMLR.
February 12, 2011 – Tim, Bob and I submitted a paper entitled Mixability is Bayes Risk Curvature Relative to Log Loss to COLT 2011.
September 20, 2010 – Javen, Tiberio, and I have put our paper on hybrid losses on the arXiv.
June 14, 2010 – Launched a new discussion site for ICML 2010.
June 9, 2010 – Gave two more talks at Yass High School as part of the Mathematicians in Schools project.
May 31, 2010 – I have made available a pre-print of my article on Generalization Bounds from the forthcoming Encyclopedia of Machine Learning.
February 16, 2010 – My paper with Bob Williamson on Convexity of Proper Composite Binary Losses was accepted to AISTATS 2010.
November 18, 2009 – Spoke about Spam filtering to Year 10 students at Yass High School as part of the Mathematicians in Schools project.
September 5, 2009 – Kernel Conditional Quantile Estimation via Reduction Revisited was accepted as a short paper to ICDM 2009.