Home
Notes
Chat
Kaggle
GitHub
YouTube
SAM
Archive
About
Home
Notes
Chat
Kaggle
GitHub
YouTube
SAM
Archive
About
Subscribe
Sign in
[P31] Architecture of TabICLv2: quantile predictions for regression
How TabICLv2 models regression with 999 conditional quantiles, pinball loss, and distribution reconstruction.
Jun 6
•
Mohit Saharan
1
Latest
Top
Discussions
[P30] Architecture of TabICLv2: many-class classification
How TabICLv2 handles classification tasks with more than 10 classes by decomposing labels on the embedding side and predictions on the output side.
Jun 6
•
Mohit Saharan
1
[P29] Architecture of TabICLv2: query-aware scalable softmax
How TabICLv2 uses query-aware scalable softmax (QASSMax) to prevent attention fading in tabular in-context learning.
Jun 5
•
Mohit Saharan
2
[P28] Architecture of TabICLv2: compression-then-ICL
How TabICLv2 compresses target-aware feature tokens into row representations, then uses dataset-wise in-context learning to predict test rows.
Jun 4
•
Mohit Saharan
3
[P27] Architecture of TabICLv2: target-aware embedding
How TabICLv2 uses target-aware embedding to add training labels to tabular in-context learning tokens while preventing label leakage in test rows.
Jun 2
•
Mohit Saharan
3
[P26] Architecture of TabICLv2: repeated feature grouping
A technical guide to TabICLv2 repeated feature grouping: why similar columns confuse encoders, how circular shifts add context, with NanoTabICL…
Jun 1
•
Mohit Saharan
3
2
2
[P25] SAM Platform: Orchestration + ML for Trading PyPI Libraries
Developing a production-oriented trading system on top of ML4T libraries.
May 25
•
Mohit Saharan
2
3
2
[P24] YouTube channel for explainer videos
Subscribe for videos about tabular foundation models, machine learning, data science, and quantitative finance.
May 22
•
Mohit Saharan
1
1
[P23] Introducing SAM: a personal quantitative research and engineering platform
A companion to the [DS, AI, Engineering] Newsletter.
May 21
•
Mohit Saharan
2
1
[P22] Tactical asset allocation with TabPFN, TabICL, and XGBoost - 3
Given information available at a monthly signal date, can TabPFN, TabICL, XGBoost, and simple allocation rules rank assets in a public ETF universe by…
May 12
•
Mohit Saharan
1
1
See all
DSAIEngineering Newsletter
Currently: machine learning, tabular foundation models, PyTorch. Pace: 220±20 workouts-turned-posts per year.
Subscribe
Recommendations
Mindful Modeler
Christoph Molnar
Social
LinkedIn
This site requires JavaScript to run correctly. Please
turn on JavaScript
or unblock scripts
