DS@GT ARC

History

The group began in Spring 2022 with participation in the BirdCLEF 2022 competition, where it won the Best Working Notes Award ($2,500 GCP credits). In Fall 2022, a preparatory “EDA” group focused on BirdCLEF data was formalized. The group continued its success in Spring 2023, participating in BirdCLEF 2023 and again winning the Best Working Notes Award ($2,500 cash).

Fall 2023 saw the establishment of the “Kaggle-CLEF Interest Group” model – a semester-long program introducing competitions (Kaggle, CLEF), fostering skills (EDA, paper reading), and preparing members for Spring teams, including an internal Kaggle competition. Spring 2024 marked a significant expansion: participation grew to 6 CLEF labs and 11 tasks with 31 members (a mix of OMSCS/OMSA, undergrad, online/on-campus). The group published 11 working notes in the CLEF 2024 proceedings (22 unique authors) and secured its third consecutive Best Working Notes Award for BirdCLEF 2024 ($2,500 cash prize). Members also presented at the OMSCS Conference and GT Showcases (PlantCLEF won the Student Choice Award), and several received course credit (CS 8903, etc.).

View this post on Instagram

A post shared by Anthony Miyaguchi (@acmiyaguchi)

In Fall 2024, the group continued the Fall Interest Group model, refining the schedule, incorporating PACE cluster training, and securing access to Georgia Tech’s PACE HPC cluster via PACESHIP alongside continued GCP use. The aim was to have over 50 participants. For Spring 2025, the group aims for participation across all CLEF labs (~15), targeting 50 active members and 15 accepted papers, and is currently on track for potentially 20 papers and 55 authors across 15 labs.