I'm attending Artificial Intelligence Festival. Next generation machine intelligence. Virtual. Attend 3 talks for free. Full access costs. RT if you think others will find this helpful.
https://t.co/awV1GSI1mi
Next theory is that the real cause is how the problem is specified. Multiple layers each have their own loss and they compete against each other, the optimizer can't find a compromise.
Here, blending weights between conv3_1 and conv1_1. Nice patterns or correct colors, pick one: