Profile avatar
bipr.bsky.social
ML & Privacy Prof at the University of Melbourne, Australia. Deputy Dean Research. Prev Microsoft Research, Berkeley EECS PhD. @bipr on the X bird site. He/him.
16 posts 1,292 followers 320 following
Prolific Poster
Conversation Starter

We are hiring continuing academic in #cyber security! Come join one of Australia's best groups (brokenassumptions.org), alongside stars like @bipr.bsky.social, @shaananc.bsky.social, Olga Ohrimenko, Thuan Pham, Xingliang Yuan, & Sarah Erfani. jobs.unimelb.edu.au/en/job/91947...

CBA trials myGov ID verification www.cyberdaily.au/security/116...

Curious to see how they plan to anonymise our NHS data for training ML models. I personally wouldn’t know how to start doing that www.theguardian.com/politics/202...

If the UK can do it, why can't Australia? www.computing.co.uk/news/2025/ai...

The NACC (Australia's Commonwealth attempt at an anti-corruption commission) has been quite the disappointment.

Excited to co-organise another summer school on learning theory! This time in beautiful Odense, Denmark. Please share and apply here: event.sdu.dk/algoml2025

BoM data finds 2024 was Australia’s second-hottest year on record www.theguardian.com/australia-ne...

I seem to be on a "defend LLMs for personal use" kick. It seemed like fun to see how I've used them in the past month. In most cases, I first did a search that proved useless (often, spending a long time following links that looked like they *could* be helpful…but weren't). ↵

I've been thinking about in-context learning for nearly 3 years. While there is still plenty I don't fully understand, five papers have--to a very large extent--shaped my perspective on it, and I believe everyone should read them.

Fuck no. www.smh.com.au/politics/fed...

📢 Outstanding PhD opportunity! The successful candidate will be based at The University of Manchester Dept. of CS ✨ to work on learning theory and methods for novel types of distributional shifts, co-supervised with @samikaski.bsky.social ⏳ DL 31.Jan.2025. www.findaphd.com/phds/project...

As the year winds down, the summer sun shines and boxing day does its thing 🏏, reminder that if you want to follow #ComputerScience academics in Australia 🇦🇺 and NZ 🇳🇿, there's a starter pack perfect for you: go.bsky.app/N5x6GLQ

Loving the new Yayoi Kusama exhibition @ National Gallery Victoria #Melbourne

Announcing the accepted papers for ALT 2025! It was a very competitive year, and we were able to accept only 51 of 144 submissions. Thanks to my co-chair Po-Ling Loh and the entire program committee for their hard work! See you in Milan in February! algorithmiclearningtheory.org/alt2025/acce...

The School of Informatics at the University of Edinburgh is hiring a lecturer or reader in embodied natural language processing. Apply by 31 Jan 2025 at edin.ac/4fqgawg

I was on NPR with @bobbyallyn.bsky.social re Australia’s teen social media ban. Tests by the US National Institute of Standards and Technology found that age algorithms are off by an average of 5yrs for West Africans–discriminating in practice. 🎧 Take a listen! www.npr.org/2024/12/09/n...

Found slides by Ankur Moitra (presented at a TCS For All event) on "How to do theoretical research." Full of great advice! My favourite: "Find the easiest problem you can't solve. The more embarrassing, the better!" Slides: drive.google.com/file/d/15VaT... TCS For all: sigact.org/tcsforall/

I look forward to co-directing the Canadian AI Safety Institute (CAISI) Research Program at CIFAR with @catherineregis.bsky.social We will be designing the program in the coming months and will soon share ways to get involved with this new community. Read more here: cifar.ca/cifarnews/20...

Now that @mkearnsphilly.bsky.social is here, this starter pack just became the real deal. Better refollow it. go.bsky.app/21nFz12

Actual content post: Have not talked much about this work yet but we have a paper on Oracle-Efficient Reinforcement Learning for Max Value Ensembles at this year's #NeurIPS. We provide an efficient algorithm to ensemble policies given a value function oracle. arxiv.org/abs/2405.16739

Touching story of the innovations of an immigrant overcoming adversity. As with so many other skilled immigrants, Australia benefits. Short sighted to scape goat people who want to build a better society, economy and world. We should welcome them instead. www.abc.net.au/news/2024-12...

An advent calendar of some of my favourite TCS/Maths talks. Day #1: Avi Wigderson on Reading Alan Turing. It is a gem of a talk, full of insights about Turing's work, writing style, and influences on mathematics and computer science. Pure joy! www.youtube.com/watch?v=_Uk_...

I'm excited about a new paper that gives tractable generalizations of Aumann's Agreement Theorem, with an eye towards human/model collaboration in machine learning. We can implement algorithms that can "converse" with people and quickly come to agreement on downstream actions. 🧵

Two weeks remaining to apply to this position. I'll also be at NeurIPS, if you want to chat you can DM or email me. :)

Test-of-time award for Asiacrypt 2009 is awarded two amazing papers on lattice-based cryptography! Big congratulations to my former PhD supervisor and now close collaborator Ron Steinfeld and other recipients including Vadim Lyubashevsky and Damien Stehle!

It's 2040. ICLR rebuttal now lasts two years. Reviewer 2 still hasn't read your paper but has strong opinions about it

Want to learn / teach RL? 
 Check out new book draft: Reinforcement Learning - Foundations
sites.google.com/view/rlfound... W/ Shie Mannor & Yishay Mansour This is a rigorous first course in RL, based on our teaching at TAU CS and Technion ECE.

The PCP theorem, a jewel of theoretical computer science, establishes that any NP statement can be assessed by a randomized verifier who only checks a vanishing fraction of the proof (indeed, a constant # of characters!) This has had incredible impact, most notably on how ML reviews are conducted

On the TCS job market: Tamalika Mukherjee! Tamalika's research: (1) designing computationally efficient privacy-preserving algorithms for data analysis, (2) understanding their societal implications through the lens of equity and policy. Keywords: sublinear, DP 1/2 #TCSSky #AcademicJobMarket

Hire Hao, he's great!

Ensemble Everything Everywhere is a defense against adversarial examples that people got quite exited about a few months ago (in particular, the defense causes "perceptually aligned" gradients just like adversarial training) Unfortunately, we show it's not robust... arxiv.org/abs/2411.14834

I noticed a lot of starter packs skewed towards faculty/industry, so I made one of just NLP & ML students: go.bsky.app/vju2ux Students do different research, go on the job market, and recruit other students. Ping me and I'll add you!

It's tough to gain visibility as a young researcher, and it's job market season! Are you a theoretical computer science PhD/postdoc on the job market? I don't have a crazy juge audience but I'll try to help a bit: fill this form, and I'll tweet your pitch and info! docs.google.com/forms/d/e/1F...

By the way, did I mention that next year's IEEE Information Theory Workshop, ITW'25, will be in Sydney? I did not? Well, it'll be in Sydney! (I'm TPC co-chair, along with Ángela Capel Cuevas and Yingbin Liang) www.ieee-itw2025.org

Help a lady out! Send me your favorite starter packs, please!

It is frankly ridiculous how the ARC toys with the time and job security of grant applicants over and over again. It demands rigid adherence to deadlines and criteria from applicants yet its own processes are vague and shifting

The Australian Government is cramming through its under-16yrs old social media ban with an insane 24 hour period for public notice and comment!!! 🤯 Kudos to my team at unimelb for putting a submission together so fast.