Haha not that different from data science in other large organizations, but from time to time you do gruard shifts because it’s the army.
People try to make you do research on problems with next to no data, and there are literal human lives on the line, but you have to convince everyone it’s not possible and you should focus on other, more approachable problems.
Let’s take, for instance, a non classified problem: Automatically detecting people trying to sneak towards you. Seemingly easy: computer vision is a solved problem! However, sneaking tends to happen at night. While wearing camouflage (Out Of Distribution!). Sometimes your cameras are on a plane (OOD!) and/ or cover a large area (very small humans!). Also, there are only a few terrorist attacks a year, only some of them properly saved (until the one that literally kills a thousand people in one day), so almost no test data, not to mention training.
In conclusion: lots of fun, lots of action, but I enjoy civilian life better. It has better data!
I feel like militaries would really want to collect huge datasets of camoflaged military personel and equipment, usually from long distance. If people collected million-image datasets and used the largest models it might be feasible
Haha not that different from data science in other large organizations, but from time to time you do gruard shifts because it’s the army.
People try to make you do research on problems with next to no data, and there are literal human lives on the line, but you have to convince everyone it’s not possible and you should focus on other, more approachable problems.
Let’s take, for instance, a non classified problem: Automatically detecting people trying to sneak towards you. Seemingly easy: computer vision is a solved problem! However, sneaking tends to happen at night. While wearing camouflage (Out Of Distribution!). Sometimes your cameras are on a plane (OOD!) and/ or cover a large area (very small humans!). Also, there are only a few terrorist attacks a year, only some of them properly saved (until the one that literally kills a thousand people in one day), so almost no test data, not to mention training.
In conclusion: lots of fun, lots of action, but I enjoy civilian life better. It has better data!
I feel like militaries would really want to collect huge datasets of camoflaged military personel and equipment, usually from long distance. If people collected million-image datasets and used the largest models it might be feasible
I imagine the US might have something like that. but if it does it doesn’t share it.