BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Ainave//Events//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VEVENT
UID:celesta-exoplanet-challenge.devpost.com
DTSTAMP:20260716T143922Z
DTSTART;VALUE=DATE:20260615
DTEND;VALUE=DATE:20260801
SUMMARY:India High School Exoplanet Data Challenge
DESCRIPTION:### Overview\nThe India High School Exoplanet Data Challenge is a public\, online competition designed for high school students across India. Participants are invited to step into the roles of astrophysicists and data scientists to tackle a real world problem using NASA Exoplanet Archive data. The goal is to build a machine learning classification model that distinguishes real exoplanet candidates from noise and false signals. The challenge emphasizes applied AI for space science\, offering a portfolio-worthy project backed by NASA data and real observational characteristics. The event is hosted on Devpost\, with a dedicated challenge page that outlines the mission\, eligibility\, deadlines\, and submission requirements.\n\n### The Mission\nYou will receive a starter CSV file containing real observational data. Your task is to explore the data\, clean it\, engineer meaningful features\, and apply classification algorithms of your choice (for example Random Forest\, XGBoost\, or Neural Networks). You should aim to produce a robust model\, accompanied by a clear explainability narrative that interprets what the model learned and why it matters. The challenge stresses accessibility: no advanced astrophysics background is required beyond curiosity\, code\, and commitment.\n\n### Who should attend\nOpen to high school students in India (grades 9–12)\, either as solo participants or in teams of up to three members. The competition is free to enter and emphasizes inclusion and learning through hands-on practice with real NASA data.\n\n### Duration and format\nThe challenge runs for about six weeks\, with a self-paced\, hands-on approach lasting roughly 1.5 months. It is designed for online participation\, with a schedule that includes a kickoff\, submission windows\, and a final presentation of results. Participants will prepare a notebook or report that documents their EDA\, data cleaning\, modeling choices\, performance metrics (accuracy\, precision\, recall\, F1)\, and an explainability narrative suitable for a non-technical audience.\n\n### Prizes and recognition\nTop entries can receive a cash equivalent prize\, Featherless.ai credits\, and opportunities such as publication in Celesta newsletters\, internships\, andLetters of Recommendation. Other top solutions may be featured in Celesta communications to the STEM community. Certificates of Excellence are awarded to all top-scoring participants\, and the challenge highlights the educational value of real NASA data for college applications and research portfolios.\n\n### Submission requirements\nParticipants must provide a reproducible notebook or Python script\, a model results summary\, a written report (500–1000 words) covering data exploration\, model choice\, key findings\, and explanations suitable for a general audience\, and a Devpost project page with embedded or linked GitHub repository and at least one visualization from the notebook. Submissions must be completed within the challenge window and in English\, with one submission per team.\n\n### Additional context\nThe event uses online channels and the Devpost platform for information\, updates\, and judging criteria. It emphasizes practical skills\, rigorous evaluation\, and the translation of complex ML results into accessible insights for a broad audience.\nhttps://celesta-exoplanet-challenge.devpost.com?ref=ainave
LOCATION:
URL;VALUE=URI:https://celesta-exoplanet-challenge.devpost.com?ref=ainave
END:VEVENT
END:VCALENDAR