Can Neural Nets Learn the Same Model Twice? Investigating Reproducibility and Double Descent from the Decision Boundary Perspective | IEEE Conference Publication | IEEE Xplore