Aishik Ghosh
@aishikghosh.bsky.social
Assistant Professor of AI in School of Physics @ GT
Fundamental Physics ∩ AI
AI Policy
AI for hypothesis generation, simulation, inference in particle & astro
High-dimensional statistics, neural inference, uncertainty quantification
Fundamental Physics ∩ AI
AI Policy
AI for hypothesis generation, simulation, inference in particle & astro
High-dimensional statistics, neural inference, uncertainty quantification
Our two new research areas are now live!
November 18, 2025 at 3:36 PM
Our two new research areas are now live!
At the launch event for @gtsciences.bsky.social's AI4Science Centre: ai4science.ai.gatech.edu. With colleagues from psychology, math, earth sciences and everything in between. Looking forward to all the cross-pollination within the sciences and also with industry!
August 26, 2025 at 10:12 PM
At the launch event for @gtsciences.bsky.social's AI4Science Centre: ai4science.ai.gatech.edu. With colleagues from psychology, math, earth sciences and everything in between. Looking forward to all the cross-pollination within the sciences and also with industry!
And we’re also listening to the constructive criticism, the most universal request, we need coffee in all the ‘coffee breaks’!
August 14, 2025 at 4:39 PM
And we’re also listening to the constructive criticism, the most universal request, we need coffee in all the ‘coffee breaks’!
A little comment like one of these makes my week. Teaching and mentorship is easily the most rewarding aspect of academia. Makes it all worth it. Thanks again to all the students who came to ML4FP and interacted with us for a week. Your positive comments mean the word to us!
August 14, 2025 at 4:39 PM
A little comment like one of these makes my week. Teaching and mentorship is easily the most rewarding aspect of academia. Makes it all worth it. Thanks again to all the students who came to ML4FP and interacted with us for a week. Your positive comments mean the word to us!
Finally, Elham E Khoda talks about efficient ML and fast inference on FPGAs
August 8, 2025 at 11:29 PM
Finally, Elham E Khoda talks about efficient ML and fast inference on FPGAs
Today we have the ATLAS, CMS and Neutrino parallel sessions at the ML4FP School. Julia Gonski give us an overview of ML in ATLAS
August 8, 2025 at 4:36 PM
Today we have the ATLAS, CMS and Neutrino parallel sessions at the ML4FP School. Julia Gonski give us an overview of ML in ATLAS
Finally, Chase Shimmin tells us about fun opportunities in the start up world!
August 8, 2025 at 4:34 PM
Finally, Chase Shimmin tells us about fun opportunities in the start up world!
Now @ozamram.bsky.social discusses simple and advanced anomaly detection techniques in HEP
August 7, 2025 at 9:14 PM
Now @ozamram.bsky.social discusses simple and advanced anomaly detection techniques in HEP
Day 4: Sascha teaching us about diffusion, flows and all things generative
August 7, 2025 at 7:06 PM
Day 4: Sascha teaching us about diffusion, flows and all things generative
Day 3: We're scratching our heads over the attention mechanism.
August 6, 2025 at 10:57 PM
Day 3: We're scratching our heads over the attention mechanism.
Day 2 we have Vinicius Mikuni and Kazuhiro Terao give overviews of AI in collider and neutrino, then an afternoon on neural simulation-based inference & uncertainty quantification with Tae Park, Sean Benevedes and I.
August 6, 2025 at 10:54 PM
Day 2 we have Vinicius Mikuni and Kazuhiro Terao give overviews of AI in collider and neutrino, then an afternoon on neural simulation-based inference & uncertainty quantification with Tae Park, Sean Benevedes and I.
Now Dennis Noll energising the room with his lecture on advanced topics in the fundamentals of ML
August 4, 2025 at 8:52 PM
Now Dennis Noll energising the room with his lecture on advanced topics in the fundamentals of ML
Kevin Greif just gave a fantastic tutorial on training your first network
August 4, 2025 at 8:50 PM
Kevin Greif just gave a fantastic tutorial on training your first network
Sascha Diefenbacher teaches us about the fundamentals of neural networks
August 4, 2025 at 8:49 PM
Sascha Diefenbacher teaches us about the fundamentals of neural networks
As you can see, we're very proud of the broad participation
August 4, 2025 at 3:53 PM
As you can see, we're very proud of the broad participation
And we're off! Yifan Chen welcomes students to the Machine Learning for Fundamental Physics School 2025!
August 4, 2025 at 3:50 PM
And we're off! Yifan Chen welcomes students to the Machine Learning for Fundamental Physics School 2025!
Bonus: Rapid Neyman construction with ML, for those tired of throwing a gazillion toys.
July 28, 2025 at 7:02 AM
Bonus: Rapid Neyman construction with ML, for those tired of throwing a gazillion toys.
Gains are quite robust to the choice of your focus function and yield valid confidence intervals by construction.
July 28, 2025 at 7:02 AM
Gains are quite robust to the choice of your focus function and yield valid confidence intervals by construction.
Works out of the box on the ATLAS HiggsML dataset and an LZ-inspired dark matter search, yielding smaller frequentist confidence intervals!
July 28, 2025 at 7:02 AM
Works out of the box on the ATLAS HiggsML dataset and an LZ-inspired dark matter search, yielding smaller frequentist confidence intervals!
The NP lemma applies to simple hypothesis tests. Searches & measurements are composite. LRS compares them pairwise, we use a test statistic that accounts for all alternates and focuses the power in physics-motivated regions
July 28, 2025 at 7:02 AM
The NP lemma applies to simple hypothesis tests. Searches & measurements are composite. LRS compares them pairwise, we use a test statistic that accounts for all alternates and focuses the power in physics-motivated regions
NSBI can be performed robustly even when there are missing neutrinos in the final state!
July 4, 2025 at 2:15 AM
NSBI can be performed robustly even when there are missing neutrinos in the final state!
We combine the advantages of matrix-element-enhanced NSBI methods that are essential for robust EFT inference with classifier-based methods that can handle different initial states.
July 4, 2025 at 2:15 AM
We combine the advantages of matrix-element-enhanced NSBI methods that are essential for robust EFT inference with classifier-based methods that can handle different initial states.
New paper: Constraining the Higgs self-coupling from the off-shell single Higgs with NSBI. Work with Tae Hyoun Park, Maximilian Griese & Ulrich Haisch.
arxiv.org/abs/2507.02032

arxiv.org/abs/2507.02032

July 4, 2025 at 2:12 AM
New paper: Constraining the Higgs self-coupling from the off-shell single Higgs with NSBI. Work with Tae Hyoun Park, Maximilian Griese & Ulrich Haisch.
arxiv.org/abs/2507.02032

arxiv.org/abs/2507.02032

Could such tools help theorists quickly test out new ideas ?
June 11, 2025 at 4:12 PM
Could such tools help theorists quickly test out new ideas ?
And lots of other unique models, full list released: github.com/jake-rudolph...
June 11, 2025 at 4:12 PM
And lots of other unique models, full list released: github.com/jake-rudolph...