Mohammad Ali Javidian

Assistant Professor in CS at Appalachian State University
Interested in Probabilistic Graphical Models and Causal Inference. I'm Persian; I'm NOT a MUSLIM.

[Mathematica @mathemetica]

Today, May 12, we celebrate the birthday of Maryam Mirzakhani (1977–2017) - the first woman and first Iranian to win the Fields Medal.

Her groundbreaking work on Riemann surfaces and moduli spaces continues to inspire mathematicians worldwide.

May 12 is now International Women in Mathematics Day.

[Goldie Ghamari | گلسا قمری 🇮🇷 @gghamari]

President Trump: "The Iranian people want to be free. They have lived in a world that you know nothing about."

Thank you President Trump. You're 100% right. The vast majority of people have no idea what Iranians have lived through.

I've never felt more seen in my entire life.

[Mohammad Ali Javidian @ali_javidian]

The Islamic regime must go; let's make Iran great again! Viva Iran, viva USA.

[RSBN 🇺🇸 @RSBNetwork]

FULL SPEECH: Prince Reza Pahlavi Addresses CPAC 2026 - 03/28/26

[Lindsey Graham @LindseyGrahamSC]

President Trump’s decision to insist on unconditional surrender of the largest state sponsor of terrorism, the Iranian Regime, which has been pursuing nuclear weapons, funding terrorist proxies, building thousands of ballistic missiles with the hope of one day reaching America, and has so much American blood on its hands, is the right and only approach to end this Iranian nightmare.

The people of Iran should not be expected to live under the remnants of a weakened terrorist regime that has slaughtered over 32,000 of its own citizens. As in World War II, the evil of Nazi Germany and Imperial Japan had to be defeated for the world to ever know peace again. The same is true with the Iranian regime. They must be defeated.

I applaud President Trump for his steadfast leadership to achieve this goal. To the men and women involved in Operation Epic Fury: your righteous efforts will lead to the collapse of one of the most evil regimes in modern times. Godspeed.

Well done, President @realDonaldTrump.

[Charlie Hebdo @Charlie_Hebdo_]

Ali Khamenei 1939-2026

👉 Dossier Municipales à Lyon : Lyon, zone tiède
👉 Khamenei : il nous manque déjà !
👉 Complotisme : les pyramides, aimants à zinzins

En vente mercredi !

[Mohammad Ali Javidian @ali_javidian]

Thank you, President Trump @POTUS @realDonaldTrump

[Open Source Intel @Osint613]

WATCH 🔴🔴

Trump addresses the nation

[Mohammad Ali Javidian @ali_javidian]

Iran's Supreme Leader Ali Khamenei has been killed in Saturday's airstrikes. Thank you, President @realDonaldTrump, we owe you big time! Thank you BiBi @netanyahu!
Congratulations to all Iranian, American, and Israeli friends. The world is a better place without Mullah.

[Mohammad Ali Javidian @ali_javidian]

Just got back from ICDM 2025 in DC and wrote a short mini-blog with my notes+reflections (keynotes, my paper/poster, and thoughts on foundations vs LLMs).
👉 https://t.co/tXpdrI9EGj
Thanks to Dr. Jay Fenwick for the support that made this trip possible.
#ICDM2025 #CausalInference

[Mohammad Ali Javidian @ali_javidian]

Overall impression: huge opportunities at the intersection of time series modeling and causal reasoning, both in theory and in practical benchmarks.

Stay tuned for more updates from ICDM 2025!
#ICDM2025 #AI4TS #TimeSeries #Causality #DeepLearning #DataScience #FoundationModels

[Mohammad Ali Javidian @ali_javidian]

🌟Day 1 Takeaways from AI4TS@ICDM 2025 in Washington, DC!🌟
I spent an incredibly stimulating day at the AI for Time Series Analysis: Theory, Algorithms, and Applications (AI4TS) workshop (https://t.co/02HOCkSMwY), diving deep into the cutting edge of Time Series (TS) modeling.

[Mohammad Ali Javidian @ali_javidian]

3️⃣ Keynotes by Dr. Papapetrou (Stockholm University) and Dr. Rex Ying (Yale University).
-Dr. Papapetrou touched on counterfactuals in time series, but interestingly, causality itself mostly appeared as “future work,” even though counterfactuals are inherently causal concepts.

[Mohammad Ali Javidian @ali_javidian]

-Dr. Ying discussed benchmarking TS models, including issues like poor alignment due to missing causal links. From my Pearl-style causality lens, the way causality entered these benchmarks still feels non-causal!

[Mohammad Ali Javidian @ali_javidian]

YouTube presentation: https://t.co/qdfqKjZ85e

[Mohammad Ali Javidian @ali_javidian]

Thrilled to share that my paper, EM-Based Transfer Learning for Gaussian Causal Models Under Covariate and Target Shift, has been accepted as a regular paper at IEEE ICDM 2025.
What’s the problem statement? When deploying models across environments, the data distribution shifts.

[Mohammad Ali Javidian @ali_javidian]

Extending to multiple latent nodes, non-Gaussian/nonlinear mechanisms, and mixed graphs (ADMGs), and analyzing robustness under structural misspecification are exciting next steps.
Resources: GitHub (code & supplementary): https://t.co/o673MXHJoF