Aaron Roth Add us as a preferred source on Google

Bluesky @aaroth.bsky.social

⚔ Compare ☆ Follow Own this channel?

🔄 Data last refreshed 2 months ago

🤝

Make a deal with Aaron Roth

🔒 Paid proposals held safely in escrow — released only when the work's approved.

Followers

4.4K

Account age

3 yrs

🎁 Free analysisfor Aaron Roth

🕵️Fake Followers CheckReal-audience % & bot risk→ 📊Engagement rateBenchmarked vs. peers→ 💰What they chargeFair sponsored-post price→

Known for

Modern LLMs are incredibly good compression algorithms, which can shed light on why autonomous data science agents don't overfit as much as you might think. arxiv.org/abs/2606.11045

28 views Modern LLMs are incredibly good compression algorithms, which can shed light on why autonomous data science agents don't overfit as much as you might think. arxiv.org/abs/2606.11045

It was fun giving this tutorial. If you missed it, check out calibration-tutorial.github.io where all of our materials are available, including slides, hundreds of pages of lecture notes, an interactive demo, and an annotated reading list.

16 views It was fun giving this tutorial. If you missed it, check out calibration-tutorial.github.io where all of our materials are available, including slides, hundreds of pages of lecture notes, an interactive demo, and an annotated reading list.

For a long time we didn't know if test-time randomization was needed for sample-optimal multicalibration and omniprediction. It's not. arxiv.org/abs/2606.20557

13 views For a long time we didn't know if test-time randomization was needed for sample-optimal multicalibration and omniprediction. It's not. arxiv.org/abs/2606.20557 $Recently we showed that the minimax optimal rate for multicalibration is T^{2/3}. But that doesn't mean you have to do that badly on all instances. We give an algorithm that can adapt to easy instances and get better rates while still being minimax optimal in the worst case. arxiv.org/abs/2605.09273$ 13 views Recently we showed that the minimax optimal rate for multicalibration is T^{2/3}. But that doesn't mean you have to do that badly on all instances. We give an algorithm that can adapt to easy instances and get better rates while still being minimax optimal in the worst case. arxiv.org/abs/2605.09273

📊 Post engagement

📊 Activity & format

Posting cadence

0.83 / week

A lower-frequency account — each post lands with more weight.

Content mix

Mostly text

Recent: 6 text · 6 image · 0 video.

Follower / following

11×

Follows 391 back. A strong ratio — an audience that follows them, not a follow-for-follow network.

Recent posts

View on Bluesky ↗

🔥 Top post People used to be able to impress and intimidate reviewers with complicated proofs. This will change. In the age of AI inscrutable proofs are cheap. It is understandable proofs that are valuable. Opaque complexity is now it is a sign of laziness or lack of insight. ★ 67

We have a new online boosting algorithm which is very efficient and effective. Unlike prior algorithms which maintain many weak learners and ensemble them, we operationalize the "dual view" of boosting. We don't maintain an ensemble. We tr… ★ 6 Right now we have "problem overhang" - lots of problems we as a community are interested in because smart and charismatic people thought about them and convinced us that these problems are important. So we are happy/interested to see them … ★ 9

It was fun giving this tutorial. If you missed it, check out calibration-tutorial.github.io where all of our materials are available, including slides, hundreds of pages of lecture notes, an interactive demo, and an annotated reading list. ★ 16 On Monday @ncollina.bsky.social @iraglobusharris.bsky.social and I are giving a tutorial at ICML on (multi)calibration and its applications. You can find slides and an annotated bibliography on the website: calibration-tutorial.github.io… ★ 48 AI is getting good at math. What are our jobs as researchers now that we have proof machines? The raw proofs that come from LLMs are difficult to understand, even if correct. So its now easy to quickly write many badly written papers that … ★ 24 Interested to see how this goes. Now that the cost of generating a paper-like-object has dropped so low, publication venues are going to start having to impose costs on submissions of various sorts. We'll need experiments to figure out th… ★ 3

For a long time we didn't know if test-time randomization was needed for sample-optimal multicalibration and omniprediction. It's not. arxiv.org/abs/2606.20557 ★ 13

Modern LLMs are incredibly good compression algorithms, which can shed light on why autonomous data science agents don't overfit as much as you might think. arxiv.org/abs/2606.11045 ★ 28

A clearly hallucinated citation! NeurIPS 2026 decisions aren't out yet. But wait --- the hallucination is also present in the bibtex entries from openreview openreview.net/forum?id=fAj... and Google Scholar scholar.googleusercontent.com/sc… ★ 2

Recently we showed that the minimax optimal rate for multicalibration is T^{2/3}. But that doesn't mean you have to do that badly on all instances. We give an algorithm that can adapt to easy instances and get better rates while still bein… ★ 13 I'm giving this talk at the MIT CS theory seminar tomorrow. Stop by if you are around! ★ 4

🦋 Handle & authenticity

Bluesky handle

@aaroth.bsky.social

A default .bsky.social handle — genuine, but not domain-verified. Confirm identity before a high-value deal.

On Bluesky since

Oct 2023

An early adopter from the invite-only beta (before the Feb 2024 public launch) — an original, established account.

💡 Facts

🗓️Joined Bluesky in 2023 — 3 years ago.

👁️Averages 19 views per post.

📤Posts about 0.8× per week.

🕵️ Fake follower check

Estimated

86/100

Excellent Credibility score

99%

Real Real audience

Low Fake-follower risk

High Data confidence

Est. 99% real, active audience · Low fake-follower risk.
Strong engagement (~0.4% of followers engage each post) — an active, real audience.
Organic base — far more followers than accounts it follows.
Established account (3+ years old).

Heuristic estimate from engagement, follower ratios, account age & growth — a screening signal, not a guarantee.

About

Professor at Penn, Amazon Scholar at AWS. Interested in machine learning, uncertainty quantification, game theory, privacy, fairness, and most of the intersections therein

📸 Gallery

$Recently we showed that the minimax optimal rate for multicalibration is T^{2/3}. But that doesn't mean you have to do that badly on all instances. We give an algorithm that can adapt to easy instances and get better rates while still being minimax optimal in the worst case. arxiv.org/abs/2605.09273$

🔀 Audience overlap

Estimated

Estimated shared audience with similar creators — useful for avoiding overlap (or doubling down) when planning a campaign.

Anne Greensmith 16%

Makili217 14%

Emile Avanessian 14%

Bonhoeffer's Child 13%

Cody Stavenhagen 13%

redwyrm 12%

More like this

Find more →

Bluesky ＋

Makili217

@makili217.bsky.social

4.4K Followers

Bluesky ＋

Anne Greensmith

@snowleopardess.bsky.social

4.4K Followers

Bluesky ＋

Emile Avanessian

@hhypeemile.bsky.social

@bonhoefferchild.bsky.social

4.4K Followers

Bluesky ＋

Cody Stavenhagen ✓

@codystavenhagen.bsky.social

4.4K Followers

Frequently asked questions

How much does Aaron Roth charge for a sponsorship?

Aaron Roth hasn't published fixed prices yet. Send a proposal through SocialDB and agree terms directly — your payment is held in escrow until the work is approved.

How do I contact Aaron Roth for a brand deal or collaboration?

Send a paid deal or collab request right here on SocialDB — we notify Aaron Roth, and once they accept, your payment is held safely in escrow and released when the work is approved. You don't need to track down an email.

How many followers does Aaron Roth have?

Aaron Roth has 4,354 followers on Bluesky.

Auto-generated from Bluesky's public data — no affiliation with or endorsement by SocialDB. Is this you? Claim it · Remove

Aaron Roth Add us as a preferred source on Google

Make a deal with Aaron Roth

Known for

📊 Post engagement

📊 Activity & format

Recent posts

🦋 Handle & authenticity

💡 Facts

🕵️ Fake follower check

About

📸 Gallery

🔀 Audience overlap

More like this

Frequently asked questions

✉ Message Aaron Roth

🎤 Event / appearance with Aaron Roth