Finally, an excuse to learn to use theorem provers and include them in my research as an auxiliary tool!
Note that they claim performance similar to TensorFlow in CPU only. But a direct translation of this to Python TensorFlow code could run in GPU, and is probably easy to make. It wouldn't be proven, but it would be much more likely to be correct than the code you have constructed for your research just now.
Also as scribu states, this doesn't allow you to prove final goals of the system like "classify images with 95% accuracy", nor does it save you from insufficient or inaccurate data.
(Edit: added last paragraph, finished reading paper)
Note that they claim performance similar to TensorFlow in CPU only. But a direct translation of this to Python TensorFlow code could run in GPU, and is probably easy to make. It wouldn't be proven, but it would be much more likely to be correct than the code you have constructed for your research just now.
Also as scribu states, this doesn't allow you to prove final goals of the system like "classify images with 95% accuracy", nor does it save you from insufficient or inaccurate data.
(Edit: added last paragraph, finished reading paper)