ℹ️ This listing was auto-generated from Mastodon's public data. Alexander hasn't claimed it, and it doesn't imply any partnership with or endorsement of SocialDB.
Is this you? Claim it · Request removal (free).
🤝
Make a deal with Alexander
🔒 Paid proposals held safely in escrow — released only when the work's approved.
Followers
56
Account age
2 yrs
💰 Suggested sponsorship rate
$40 – $65 / sponsored post
Estimated from audience size & typical reach — every deal is negotiable and held in SocialDB escrow.
🧰 Free analysis for Alexander
Known for
2 views
... the performance of established causal discovery algorithms if R²-sortability is high, as is the case on many simulated ANMs. This opens up exciting new questions:
2 views
... R² is scale-invariant, so it won't be affected by arbitrary data scales or standardization.
We show that this ordering-pattern is prevalent in many ANMs (especially with scale-free graphs as shown below), and give a sufficient criterion for causal chains.
2 views
... along the causal order if effect sizes are strong and noise variances are drawn iid.
We can exploit this and sort variables by their R² given all other variables. We introduce “R²-sortability” to measure the agreement between the R² ordering and a causal order. Importantly...
📊 Post engagement
2
Avg engagement / post
3.6%
Engagement vs followers
Aug 2023
On Mastodon since
🔥 Top post: We have developed a new Scale-Invariant Causal Discovery Baselin · 5 likes + reposts
📊 Activity & format
Posting cadence
0.09 / week
A lower-frequency account — each post lands with more weight.
Content mix
Mostly text
Recent: 8 text · 3 image · 0 video.
Follower / following
0.1×
Follows 457 back. A more reciprocal / networked account.
Recent posts
View on Mastodon ↗
Fascinating take contrasting mathematics as a theorem factory versus as a "collective endeavor focused on intelligibility", and the relevance of that distinction in the age of AI:
https://davidbessis.substack.com/p/the-fall-of-the-theorem-…
Our new article is now on arXiv:
**The Case for Time in Causal DAGs**
https://doi.org/10.48550/arXiv.2501.19311
We propose an explicit notion of time for the variables in causal DAGs and argue that this is essential for interpreting causal…
Not the scientific revolution we needed, but the scientific revolution we deserved.
How have I never heard about this?
Really cool article showing how a pseudo "Dunning-Kruger effect" can seem to show up in the data even if there is no relationship between skill and self-assessment whatsoever 😮
https://economicsfromtheto…
... our library CausalDisco (https://causaldisco.github.io/CausalDisco/) to
1. try out the R²-SortnRegress baseline, and
2. measure the R²-sortability of your own data.
Many thanks to my awesome co-authors and supervisors Myriam Tami, @chr…
- How good is your favorite algorithm compared to the R²-SortnRegress baseline?
- What R²-sortability do we find on real-world data?
- How do we best simulate ANMs that aren't close to deterministic?
Have a look at...
... the performance of established causal discovery algorithms if R²-sortability is high, as is the case on many simulated ANMs. This opens up exciting new questions:
When R²-sortability is high, as is often the case, we can exploit it using the simple “R²-sortability” algorithm:
1. sort by increasing R², and
2. regress each variable onto its predecessor in the resulting order.
This algorithm can match …
... R² is scale-invariant, so it won't be affected by arbitrary data scales or standardization.
We show that this ordering-pattern is prevalent in many ANMs (especially with scale-free graphs as shown below), and give a sufficient criterio…
... along the causal order if effect sizes are strong and noise variances are drawn iid.
We can exploit this and sort variables by their R² given all other variables. We introduce “R²-sortability” to measure the agreement between the R² or…
... the performance of established causal discovery algorithms if R²-sortability is high, as is the case on many simulated ANMs. This opens up exciting new questions:
When R²-sortability is high, as is often the case, we can exploit it using the simple “R²-sortability” algorithm:
1. sort by increasing R², and
2. regress each variable onto its predecessor in the resulting order.
This algorithm can match …
... R² is scale-invariant, so it won't be affected by arbitrary data scales or standardization.
We show that this ordering-pattern is prevalent in many ANMs (especially with scale-free graphs as shown below), and give a sufficient criterio…
... along the causal order if effect sizes are strong and noise variances are drawn iid.
We can exploit this and sort variables by their R² given all other variables. We introduce “R²-sortability” to measure the agreement between the R² or…
🐘 Community & instance
Home server
mathstodon.xyz
Their home server on the fediverse — the instance a creator picks signals the community they belong to.
On Mastodon since
Aug 2023
An established account with real history on the platform.
💡 Facts
🗓️Joined Mastodon in 2023 — 2 years ago.
👁️Averages 2 views per post.
📤Posts about 0.1× per week.
🕵️ Fake follower check
Estimated75/100
Good
Credibility score
92%
Real
Real audience
Low
Fake-follower risk
High
Data confidence
- Est. 92% real, active audience · Low fake-follower risk.
- Strong engagement (~3.6% of followers engage each post) — an active, real audience.
Heuristic estimate from engagement, follower ratios, account age & growth — a screening signal, not a guarantee.
About
Causality & Machine Learning
📸 Gallery
🔀 Audience overlap
EstimatedEstimated shared audience with similar creators — useful for avoiding overlap (or doubling down) when planning a campaign.
Paul MacLellan
16%
Al McKinlay
15%
Chris
14%
Ivana Daskalovic
12%
james makes games
12%
More like this
Find more →
Mastodon
Ivana Daskalovic ✓
@ivana@infosec.exchange
56
Followers
Mastodon
__Miguel_
@nanianmichaels@tech.lgbt
56
Followers
Mastodon
james makes games ✓
@artofjames@mastodon.gamedev.place
56
Followers
Mastodon
Paul MacLellan
@insidecircles@mastodon.scot
56
Followers
Mastodon
Al McKinlay
@thescotbot@mastodon.scot
56
Followers
Mastodon
Chris
@chris4man@mastodonapp.uk
56
Followers
✉ Message Alexander
Reaching out to influencers is a Pro feature. Upgrade to message any influencer directly — perfect for brands and agencies booking sponsorships.
- ✓ Message any influencer from their listing
- ✓ The influencer gets notified by email
- ✓ Manage every conversation in one inbox
Already Pro? Log in.
🎤 Event / appearance with Alexander
Booking an event / appearance is a Pro feature. Upgrade to book Alexander for an in-person or virtual appearance — payment held safely in escrow until the event is done.
- ✓ Book them for events, livestreams, panels & more
- ✓ Alexander gets notified by email
- ✓ Fee held in escrow, released after the appearance
Already Pro? Log in.
🚀
You're out of free requests this month
Free accounts get 5 per month. Go Pro for unlimited sponsor pitches, collab requests & sponsorship deals — plus featured placement, the Verified badge, free withdrawals and more.
Upgrade to Pro — $9.95/mo →Your free limit resets on the 1st of next month.