On the alpha-loss Landscape in the Logistic Model

06/22/2020
by   Tyler Sypherd, et al.
0

We analyze the optimization landscape of a recently introduced tunable class of loss functions called α-loss, α∈ (0,∞], in the logistic model. This family encapsulates the exponential loss (α = 1/2), the log-loss (α = 1), and the 0-1 loss (α = ∞) and contains compelling properties that enable the practitioner to discern among a host of operating conditions relevant to emerging learning methods. Specifically, we study the evolution of the optimization landscape of α-loss with respect to α using tools drawn from the study of strictly-locally-quasi-convex functions in addition to geometric techniques. We interpret these results in terms of optimization complexity via normalized gradient descent.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset