Alexander Terenin - on the faculty job market
@avt_im
Followers
8K
Following
2
Media
148
Statuses
3K
Decision-making under uncertainty, machine learning, artificial intelligence, from theory to practice · anti-ideological · Assistant Research Professor @Cornell
Joined August 2013
Paper update: our recent work on Thompson sampling has a shiny new - and I hope much better - name! This new name does much better job of emphasizing what we actually do. Joint work with @jeffNegrea. Thread below!
4
9
77
An LLM-generated paper is in the top 17% of ICLR submissions in terms of average reviewer score, having received two 8's. The paper has tons of BS jargon and hallucinated references. Fortunately, one reviewer actually looked at the paper and gave it a zero. 1/3
28
90
988
I am Socrates. But a Chad. The only thing I know is that I know nothing. I ask questions. I seek knowledge. I am here to learn.
2
1
52
This year, ICML will publish the submitted version of a paper, not just the accepted one.
- New guidelines on generative AI considerations Check out the full CfPs! Papers: https://t.co/4ppHEb6w1c Position Papers: https://t.co/HS6AXFehDW
1
5
43
The slides from our INFORMS tutorial on "The Gittins Index: A Design Principle for Decision-making Under Uncertainty" - specifically for my part - are now online! If you're interested - check them out - link below.
1
0
26
Where can we find meaning in today's world? Now available on the Forever Young Substack: How To Ensure A Longer Life Is Also A Happier One- featuring renowned psychiatrist Dr. Thomas Lewis. Link in bio.
0
0
2
@NoThanksHoney1 No, that's the mistake. People are distracted by marketing jargon like “intelligence” and “learning,” and dreams about replacing cognition. Meanwhile, what we really did was wire together two boring yet powerful technologies: statistics & (fast) arithmetic. That's a change that
3
3
11
Today, I gave a talk at the INFORMS Job Market Showcase! If you're interested, here are the slides - link below!
3
15
126
Pretty crazy that point convergence of Nesterov was open for 40+ years. I think the takeaway is that research requiring human calculations are often very suboptimal, and AI assistance can significantly improve the process.
I used ChatGPT to solve an open problem in convex optimization. *Part III* 1/N https://t.co/6HDAr6y8Z9
2
2
77
I am hiring a fully-funded #PhD in #ML to work at @EdinburghUni on 𝐠𝐞𝐨𝐦𝐞𝐭𝐫𝐢𝐜 𝐥𝐞𝐚𝐫𝐧𝐢𝐧𝐠 and 𝐮𝐧𝐜𝐞𝐫𝐭𝐚𝐢𝐧𝐭𝐲 𝐪𝐮𝐚𝐧𝐭𝐢𝐟𝐢𝐜𝐚𝐭𝐢𝐨𝐧. Application deadline: 31 Dec '25. Starts May/Sep '26. Details in the reply. Pls RT and share with anyone interested!
4
8
31
This memory lane tour reminded me of a few anecdotes. Here is a quirkier, somewhat nostalgic, alternative path through the history of AI. We held a CIFAR workshop at @UofT to learn about how to use GPUs for Deep Learning. We learned a lot from @npinto. Soon after we had the
This is an excellent history of LLMs, doesn't miss seminal papers I know. Reminds you we're standing on the shoulders of giants, and giants are still being born today.
14
14
134
Jascha's machine learning research has consistently been high-quality, interesting, and in certain cases well ahead of its time - over many years and different topics. I encourage everyone to look at his advice for researchers in today's era!
Title: Advice for a young investigator in the first and last days of the Anthropocene Abstract: Within just a few years, it is likely that we will create AI systems that outperform the best humans on all intellectual tasks. This will have implications for your research and
0
3
21
Check out the work at: https://t.co/2dcnCGTzKV And, again, shoutout to amazing coauthor @jeffNegrea! Working together has been a great pleasure! Stay tuned for follow-up: we've been working on using this viewpoint to understand other correlated perturbation-based algorithms.
arxiv.org
We develop a form Thompson sampling for online learning under full feedback - also known as prediction with expert advice - where the learner's prior is defined over the space of an adversary's...
1
1
6
So this sums up the work! If you followed along, thanks for the interest! I think you'd agree that "Bayesian Algorithms for Adversarial Online Learning: from Finite to Infinite Action Spaces" is a much better title than before. The old one was much harder to pronounce.
1
0
4
Zenith Fund high frequency trading, near riskless arbitrage capitalizes on structural inefficiencies in digital markets. 1st tranche netted 99% using multi-exchange programmatic trading resolving the time-risk challenge maintaining optimized asset balances in real time.
0
0
9
The Bayesian viewpoint proves useful for developing this analysis. It allows us to guess what a good prior will be, and suggests ways to use probability as a tool to prove the algorithm works.
1
0
1
We prove that the Bayesian approach works in this setting too. To achieve this, we develop a new probabilistic analysis of correlated Gaussian follow-the-perturbed-leader algorithms, of which ours is a special case. This has been an open challenge in the area.
1
0
1
The second one is where X = [0,1]^d and Y is the space of bounded Lipschitz functions. Here, you can't use a prior with independence across actions. You need to share information between actions. We do this by using a Gaussian process, with correlations between actions.
1
0
1
The first one is the classical discrete setting where standard algorithms such as exponential weights are studied. You can a Gaussian prior which is independent across actions.
1
0
1
Okay, so we now know what "Bayesian Algorithms for Adversarial Online Learning" are. What about "from Finite to Infinite Action Spaces"? This covers the two settings we show the aforementioned results in.
1
0
1
This approach appears to not make any sense: the Bayesian model is completely fake. We're pretending to know a distribution for how the adversary will act in the future. But, in reality, they can do anything. And yet... we show that this works!
1
0
2
We show that this game secretly has a natural Bayesian strategy - one we show is strong. What's the strategy? It's really simple: - Place a prior distribution of what the adversary will do in the future - Condition on what the adversary has done - Sample from the posterior
1
0
2