Following its success for vision and text, the "foundation model" (FM) paradigm -- pretraining large models on massive data, then fine-tuning on target tasks -- has rapidly expanded to domains in the sciences, engineering, healthcare, and beyond. Has this achieved what the original FMs accomplished, i.e. the supplanting of traditional supervised learning in their domains? To answer we look at three modalities -- genomics, satellite imaging, and time series -- with multiple recent FMs and compare them to a standard supervised learning workflow: model development, hyperparameter tuning, and training, all using only data from the target task. Across these three specialized domains, we find that it is consistently possible to train simple supervised models -- no more complicated than a lightly modified wide ResNet or UNet -- that match or even outperform the latest foundation models. Our work demonstrates that the benefits of large-scale pretraining have yet to be realized in many specialized areas, reinforces the need to compare new FMs to strong, well-tuned baselines, and introduces two new, easy-to-use, open-source, and automated workflows for doing so.

The world is in a climate crisis. With average global temperatures increasing every year, the threat of seasonal forest fires is becoming increasingly worse. In places like the Pacific Northwest, wildfire season causes extensive damage to woodlands, rural communities, and townships, destroying farmlands and infrastructure and forcing hundreds of thousands of residents to flee their homes. These fires also lead to terrible air quality in cities located hundreds (or even thousands) of miles away. For instance, in September of 2022, the city of Vancouver (British Columbia) was ranked as having the worst air quality in the world - per the Air Quality Index (AQI).

Ag-tech firm Indigo Ag has bought a company that uses satellites and artificial intelligence to operate a "living map'' of the world's food supply, Indigo …

Instead, Descartes relies on 4 petabytes of satellite imaging data and a machine learning algorithm to figure out how healthy the corn crop is from space. Grain elevator operators, ethanol producers, commodities traders, hedge funds, insurance companies, and even the farmers growing the corn will all look to the USDA's August crop report being released August 12th to try and understand how the supply side of the corn market will behave. Descartes says it can consistently out-predict the USDA's corn estimates Descartes, which launched in 2014, began releasing corn yield estimates ahead of the USDA's August crop report last year. Now new nanosatellite constellations, like the one run by satellite imaging startup Planet, are taking snapshots of the entire globe at 3- to 5-meter resolutions every day.