You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi, i have a doubt about the final scores of proposals from sliding window. It's said to be
(p_t(a_k)·p_c(a_k)) , but the p_t(a_k) from PATE indicating the trustworthiness of TAG is hoped to be lower than a threshold(0.1). Thus, the result of multiplication may be a pretty low score. Did I understand something wrong?
The text was updated successfully, but these errors were encountered:
Hi, i have a doubt about the final scores of proposals from sliding window. It's said to be
(p_t(a_k)·p_c(a_k)) , but the p_t(a_k) from PATE indicating the trustworthiness of TAG is hoped to be lower than a threshold(0.1). Thus, the result of multiplication may be a pretty low score. Did I understand something wrong?
The text was updated successfully, but these errors were encountered: