Benjamin RubinsteinBenjamin Rubinstein
Senior Lecturer (equiv US Assistant Professor)
School of Computing & Information Systems
The University of Melbourne, Australia 

Email: [firstname].[lastname]
Office: Room 7.21 Doug McDonell Building

Associate Investigator, ARC Centre of Excellence for Mathematical & Statistical Frontiers [link]
Academic Fellow, Centre for Business Analytics, Melbourne Business School [link]

LinkedIn  Twitter  GitHub   Scholar  arXiv  ORCID  Find an Expert

Juliet & Ella   Liam   Lachlan

Interests: Statistical Machine Learning, Security & Privacy, Databases, Industry Engagement

Bio: Ben joined the University of Melbourne in 2013 as a Senior Lecturer in Computing and Information Systems (a RAMAP appointee). Previously he gained four years of industry experience in the research divisions of Microsoft, Google, Intel and Yahoo! (all in the US); followed by a short stint at IBM Research Australia. As a full-time Researcher at Microsoft Research, Silicon Valley, Ben shipped production systems for entity resolution in Bing and the Xbox360 (driving huge success accounting for revenues in the $100m's); his research has helped identify and plug side-channel attacks against the popular Firefox browser, and deanonymise an unprecedented Australian Medicare data release, prompting introduction of the Re-identification Offence Bill 2016. He actively researches topics in machine learning, security & privacy, databases such as adversarial learning, differential privacy and record linkage respectively. His work has been recognised through an Australian Research Council DECRA award, and a Young Tall Poppy Science award. Ben earned the PhD in Computer Science from UC Berkeley under Peter Bartlett in 2010, collaborating closely with the SecML group, at the boundary of machine learning and security.



At the University of Melbourne since Oct'2013, awarded $2.1m in funding of which $1.94m as lead-CI; and $1.06m on a per-CI basis. Funding includes:

Other News

  • 10/2017: Invited speaker at the DARPA Safe ML workshop at the Simons Institute, Berkeley.
  • 7/2017: Open-source R diffpriv package released on CRAN based on ICML'2017 and AAAI'2017 papers. Provides many generic mechanisms for differential privacy, along with a sensitivity sampler that privatises arbitrary functions under random differential privacy, without the need to bound sensitivity or perform any math analysis whatsoever.
  • 7/2017: PhD student Maryam Fanaeepour wins the WiE Best Postgraduate Paper Award from the IEEE Australia Council for our ICDM'16 paper.
  • 3/2017: Open-source python OASIS package released based on VLDB'2017 paper to PyPI. Provides a principled, light-weight system for sampling examples to annotate for efficient (low budget) evaluation of highly class-imbalanced problems such as in record linkage. Implements this paper with Neil Marchant.
  • 1/2017: Media coverage of our liver transplantation prediction project 9news, heraldsun
  • 1/2017: Speaker at the AMIRA Exploration Managers Conference, RACV Club Healesville, March 2017.
  • 11/2016: Speaker/panelist at the public lecture "Human and Machine Judgement and Interaction Symposium", Uni Melbourne, Nov 21st
  • 11/2016: Victorian Young Tall Poppy Science Awards 2016, awardee
  • 10/2016: Deanonymisation of unprecedented public release of Australian Medicare data (30yrs of unit records for 10% of population), with Chris Culnane and Vanessa Teague. Announcement article and Dept Health press release; proposed legislation criminalising deanonymisation by Attorney-General, then subsequent exemption of White Hats; confirmation of Australia's Privacy Commissioner's acting position and his statement. Additional reporting and statements: zdnet (again), The Register, itnews (again), ABC news, The Guardian, The AgeCSO, HuffPo, Canberra Times, CrickeyComputerWorldGizmodoDigital Rights Watch, The Saturday Paper
  • Our response to the bill amending the Privacy Act to criminalised reidentification. Our submission to the Parliamentary Inquiry. Our op-ed in the SMH.
  • 05/2016: Invited speaker at the National Fintech Cyber Security Summit at the Ivy, Sydney hosted by Data61, Stone & Chalk, the Chief Scientist of Australia.
  • 04/2016: Speaking at Telstra (data science) and again 01/2017
  • 02/2016: Speaking at Samsung Research America and UC Berkeley.
  • 02/2016: Speaking in two exciting panels at AAAI'2016 on keeping AI beneficial and challenges for AI in cyber operations.
  • 12/2015: Plenary at the 12th Engineering Mathematics and Applications Conference (EMAC'2015) the biennial meeting of the EMG special interest group of ANZIAM
  • 07/2015: Keynote at the Australian Academy of Science Elizabeth and Frederick White Research Conference on Mining Data for Detection and Prediction of Failure in Geomaterials [link]
  • 12/2014: Excellence in Research Award 2014, Dept CIS, University of Melbourne
  • 11/2014: Facebook (Menlo Park) talk Data Integration through the Lens of Statistical Learning

Research Group

  • Postdocs
    • Shen Wang (2016 - ). Probabilistic programming, databases, record linkage.
    • Yi Han (2016 - ). Adversarial machine learning.
  • PhD students
    • Lingjuan Lyu (w Marimuthu Palaniswami 2016 - ). Privacy in distributed sensing.
    • Neil Marchant (w Aurore Delaigle 2016 - ). Adaptive sampling.
    • Yuan Li (w Trevor Cohn 2015 - ). Bayesian optimisation, NLP.
    • Xunyun Liu (w Raj Buyya, Rodrigo Calheiros 2015 - ). Stream computing.
    • Safiollah Heidari (w Raj Buyya 2015 - ). ML for distributed computing.
    • Yamuna Kankanige (w James Bailey 2015 - ). Liver transplant outcomes, Austin Health
    • Zay Aye - CompSci (w Rao Kotagiri 2014 - ). Learning distance metrics.
  • PhD students - Completed
    • Maryam Fanaeepour - CompSci (w Lars Kulik, Egemen Tanin 2014 - 17). Location data privacy → Postdoc at Duke University
    • Jiazhen He - CompSci (w James Bailey, Rui Zhang 2014 - 17). Education analytics and MOOCs → Postdoc at Univ Melbourne
    • Zuhe Zhang - Maths & Stats (w Sanming Zhou 2014 - 17). Differential privacy in Bayesian statistics → Data Scientist at ANZ Bank
  • Masters students
    • Completed: Rui Hu, Justin Liang, Nouras Fatima, Zhe Lim (2014); Ben Schroeter, Soundarya Mallemarapu (2015); Samuel Jenkins, Xianjing Fan, Si Chen (2016); Zean Qin, Siyu Feng, Dongge Liu, Tolga Ozdogan, Xianzhuo Ren (2017)
  • Interns from Microsoft Research


I co-designed several exciting programs offered by my department (within Engineering): the Master of Business Analytics (joint with the Melbourne Business School), the Bachelor of Science major in Data Science and the Master of Data Science (both joint with Maths & Stats within Science). These programs reflect the interdisciplinary nature of machine learning and its allied areas.












Pre 2007