Machine Learning for Planetary Science and Space Physics (ML4PSP) Seminar Series 2026 — Relaunch Announcement
The Machine Learning for Planetary Science and Planetary Exploration (ML4PSP) Seminar Series is back for 2026. This seminar series aims to bring together researchers in planetary science, space physics, machine learning, and other domains of data science. From hyperspectral data analysis to foundation-model applications on planetary datasets, ML4PSP highlights emerging methods that are shaping the future of planetary exploration. We welcome researchers, students, and practitioners from across disciplines to join, learn, and engage with this growing community.
Upcoming Seminar
Speaker: Jichao Fang (Northern Illinois University)
Title: Planetary-Scale Similarity Search for Mars Orbital Imagery with Foundation-Model Embeddings
Date: April 21, 2026
Time: 9:00 AM Pacific Time
Abstract:
Mars orbital archives now contain enough imagery that finding morphologically similar features is bottlenecked by search, not data. We present a planetary-scale similarity search system built on foundation-model embeddings over the full CTX Murray global mosaic (~26.9M indexed locations). A Vision Transformer pretrained via self-supervised learning on millions of CTX patches produces embeddings that capture surface texture and landform semantics without any labels. Deployed as a quantized vector index on a single server, the system supports sub-second instance-level retrieval (“find terrains like this”), geo-filtered search within regions of interest, and interactive relevance feedback for iterative refinement. The system is publicly accessible at findmars.space.
Join the seminar
Zoom Link: https://berkeley.zoom.us/j/93560880593?pwd=n9bfkFSrGx1A0rPKjNhx7azV532rK6.1
Meeting ID: 935 6088 0593
Passcode: ml4psp
Full schedule and details: https://ml4psp.github.io/
Organized by the ML4PSP Team
Ramana Sankar, Dona Kuruppuaratchi, Indhu Varatharajan