diff --git a/README.md b/README.md index b331d85..b8c648c 100644 --- a/README.md +++ b/README.md @@ -76,7 +76,7 @@ This is a testing sandbox for developing various methods of injecting symbolic k ```math Loss = ( \alpha * { r }^2 + 1) * SmoothL1Loss ``` - - Final adaptive semantic loss function tested was the following: + - Final adaptive semantic loss function tested (called `positive_slope_linear_loss`) was the following: ```math Loss = ( \alpha * { r }^2 + 1) * ( \frac{ 1 }{ \beta } * log( 1 + exp( \beta * \gamma * -m ) ) + 1 ) * SmoothL1Loss ```