3 ms·
A quick way out of this issue is to ask users not for how much to stake, but how confidently they believe something to be true, and then reward/penalise based o
by kqr 4y ago
A quick way out of this issue is to ask users not for how much to stake, but how confidently they believe something to be true, and then reward/penalise based on a score function derived from this and the actual outcome.