Your question is Training With Wet-Lab Uncertainty. Take a moment with it on the right.
Talk me through your thinking if you like. When you're confident, submit your answer and I'll grade it like a real screen (7/10 or better passes).
If a wet-lab team provides you with inconsistent experimental data, how do you adjust your training process or loss functions to account for this uncertainty at Flagship ventures?