Will Alexplainable recognize at least one class with at least 70% accuracy, with human guidance?
7
Ṁ180Ṁ268Jun 6
85%
chance
1H
6H
1D
1W
1M
ALL
Will Alexplainable construct a system that:
Can recognize one class (that we can choose as we want) of ImageNet with at least 70% accuracy
The system can be explained in terms of how it is mapped out and what part does what
The system does not use more than 3 layers, ignoring maxpooling and other non-complex types
Any amount of human guidance and intervention is allowed
This question is managed and resolved by Manifold.
Market context
Get 
1,000 to start trading!
1,000Sort by:
Post is now UP!
We could resolve this to YES, however it's unclear the features trained on their segments actually behave by recognizing the actual segments.
After more considerations, we will not for now continue on this project much before funding, so all markets might N/A, except maybe this one