Peter Vamplew
@amp1874
Professor in IT @ Federation Uni. Multi-objective reinforcement learning. Human-aligned AI. Best known for the f*cking mailing list paper. Jambo & Bengals fan.
I spotted this near the checkout at Woolies last night. As a society, we are so screwed :-(
I spotted this rare sighting a few years ago - a wild cellular automaton in it's natural habitat.
One of the best days of my life. Me and Zico leading out Hearts at Tynecastle in 1983.
The rise in Deep RL since around 2015 coincided with diminished interest in on-policy methods such as Sarsa and Expected Sarsa relative to off-policy approaches like Q-learning. This figure shows an analysis of reinforcement learning publications since 1990 (statistics derived from Google Scholar).
The Age's AI generated world cup 'simulation' makes no sense at all. There's no simulation here, just hallucinations that defy mathematics eg Czechia is 3 times more likely to win it all than to qualify for the semi-finals! www.theage.com.au/sport/soccer...
John Bello's profile seems normal, listing multiple publications under their name. But if you download the pdf, it's actually listing Yahaya as the sole author.
The Guardian's tool for searching the budget doesn't even list Education (or Science or Research) as an option.
A practice run for my role as mace bearer at tomorrow's graduation ceremony.
I am a bit disappointed in the reference information which the AI has generated for our Practical Guide paper. Ann Nowe is a fantastic researcher and contributed a lot to our paper, but I still don't think she deserves to be given my authorship slot in addition to her own! 😊
Everyone ran a PB? Are we sure this isn't the athletic track from The Games?
This is a hell of a coincidence. All morning I've had the song Ma Baker by Boney M stuck in my head. Now I check my bsky notifications, and someone called Ma Baker is following me?
I'm not a fan of some of terms used in this review form from IEEE Transactions on Emerging Topics in Computational Intelligence. Using words like "useless" and "dreadful" is unlikely to encourage the sort of constructive, respectful feedback that reviewers *should* be giving.
This analysis of trends on arxiv shows the same as my previous evaluation of Google Scholar publications. Multi-objective reinforcement learning has seen huge growth as a research area in recent years. In particular 2025 saw a real spike in activity on this topic.
This time I'm more confident that claim will hold, as MORL rose to 1.46% of RL papers in 2025 from 0.83% in 2024- partly due to the growth in MORL, and partly due to an apparent downward trend in RL publications since 2023 (see chart below - I'm not sure how much I trust Scholar's statistics here).
Also, thanks to the vagaries of Google Scholar's counting of RL papers, this marked the 3rd (or maybe 4th?) year in a row where I have been able to claim that this was the first time that MORL papers exceed 1% of all RL papers!
It's a new year, and that means it's time for my annual quantitative review of multi-objective reinforcement learning research. Yet again this year saw a substantial increase in the number of papers mentioning MORL, exceeding 2000 for the first time ever. 1/3
We were playing Flamecraft last night (great game btw), and I got the Flying Purple Meeple Peter.
We spotted two of these bluebottles washed up on the shore in Warrnambool today.
I'm feeling much more organised after tidying up my office yesterday. I finally have all my ducks in a row.
@tresvillain.bsky.social Andrew, for consistency you need to change your name to Trendvillian :-)
Is ResearchGate's user base dying? Reads and Full-text Reads of my papers peaked back in 2021, and have fallen by about 50% since then. Citations of my work have continued to rise, so I don't think this is caused by people losing interest in my work :-)