I'm excited to present our new method for enhancing knowledge discovery from tabular data using tabular foundation models at the #ECAI2025 conference workshops (ANSyA) this Sunday, October 26th.

See our paper, Discovering Association Rules in High-Dimensional Small Tabular Data, below 👇.

🧵1/2

📊 On 5 real-world gene expression datasets with 18K+ columns and <100 rows, both methods led to significantly higher quality association rules in terms of confidence and association strength, with limited increase in execution time (1.2 - 2 times at most).

🧵7/8

1️⃣ Aerial+WI. Weight initialization of Aerial+ based on tabular data embeddings from a foundation model, using a projection encoder.

2️⃣ Aerial+DL. Tabular embeddings are aligned with Aerial+'s reconstructions via a projection encoder and joint loss, ensuring a better semantic column alignment.

🧵6/8

🐍 Library: github.com/DiTEC-projec...

📊 We show that knowledge discovery from high-dimensional tables, as in gene expression datasets (~18K columns), is scalable with Neurosymbolic rule learning, Aerial+, a method we have proposed earlier (arxiv.org/pdf/2504.19354, presented at NeSy 2025).

🧵2/8

🎯 What can it do (technical)?

In just 2 lines of Python code, PyAerial can learn a concise set of high-quality association rules from a table in pandas dataframe form, by utilizing an under-complete denoising autoencoder!

🧵5/8

New software paper published in the SoftwareX journal, describing our neurosymbolic scalable association rule learner PyAerial for knowledge discovery and interpretable inference!

🔎 See the practical implications, capabilities, and benchmarking below!

📜 lnkd.in/eQubA7MD
🐍 lnkd.in/eDWnVWFr

🧵1/8

Tomorrow at the Neurosymbolic Learning and Reasoning (NeSy2025) conference, I will be presenting our novel, scalable association rule 'learning' method for tabular data.

Say hi if you are interested in knowledge discovery and/or interpretable ML!

📜 tinyurl.com/48fmu3eh
🐍 tinyurl.com/45s75r6w

✅ Aerial+ Rule extraction

If a forward run on the trained model with a set of marked categories A results in successful reconstruction (high probability) of categories C, we say that marked features A imply the successfully reconstructed features C, such that A → C \ A (no self-implication).

🧵5/8

✅ Method: Aerial+

Learn a compact neural representation of the tabular data using an under-complete denoising Autoencoder.

Extract association rules from the trained Autoencoder by exploiting its reconstruction mechanism

🧵4/8

Rule extraction.

After training, if a forward run on the trained model with a set of marked features (of interest) 'A' results in successful reconstruction (high probability) of features 'C', we say that marked features A imply the successfully reconstructed features C, such that A → C.

🧵4/5

Our method Aerial+ works in 2 stages:

1. It creates a neural representation of the tabular data using an under-complete denoising Autoencoder trained for rule mining.
2. Then, it extracts association rules from the neural representation by exploiting Autoencoder's reconstruction mechanism.

🧵2/5

On Wednesday (16th of April) at the NWO ICT.Open, I will present our work on Interpretable Decision-Making in (Digital Twins of) Smart Environments for high-stakes decision-making.

If you are interested, please visit me on the exhibition floor!

See the details below 👇