Here is a recap of the NASA Ames AI-Astrobiology initiative inaugural year:

• Kicked off new seminar series
• Expanded access to AI/ML literature and learning resources
• Selected to host foundation models workshop

SEMINARS

The AI-ML Seminar Series is off to a great start. Recent topics include convolutional neural networks (CNN's), machine learning and biogenicity, autonomy in space missions, and unsupervised/supervised learning. All talks available on the AI-ML Astrobiology YouTube channel.

Talks coming in the new year:

• Auto encoders
• AI and validation
• Self-assessment
• Foundation models
• Mars rovers

WORKSHOP NEWS

The AI-ML team at NASA Ames (Ryan and Caleb) has teamed up with the SETI Institute to host a workshop on Astrobiology and Foundation Models in February 2025. Applications and invites are being circulated amongst the astrobiology community and data science industry. Findings will be published after the workshop, so stay tuned!

DID YOU KNOW?

A Foundation Model (FM) refers to a diverse class of very large, AI/ML models that are trained on broad, unlabeled data and can be readily used as a basis for creating more specialized models efficiently. FMs can use techniques such as self-supervision (a form

of unsupervised learning) and many are built around deep-learning neural networks. An example of an FM, well known to the public, is the large-language model (LLM) GPT.

Try a Machine Learning Tutorial!

HIGHLIGHTS

Science in the age of AI

AI in-situ exploration - Dr. Bethany Theiling

Two-part education series - Dr. Floyd Nichols

TelescopeML and Convolutional Neural Networks – Dr. Ehsan “Sam” Gharib-Nezad