A Unified and Fast Interpretable Model for Predictive Analytics

11/16/2021
by   Rui Ding, et al.
0

In this paper, we propose FXAM (Fast and eXplainable Additive Model), a unified and fast interpretable model for predictive analytics. FXAM extends GAM's (Generalized Additive Model) modeling capability with a unified additive model for numerical, categorical, and temporal features. FXAM conducts a novel training procedure called Three-Stage Iteration (TSI). The three stages correspond to learning over numerical, categorical and temporal features respectively. Each stage learns a local optimum by fixing parameters of other stages. We design joint learning over categorical features and partial learning over temporal features to achieve high accuracy and training efficiency. We prove that TSI is guaranteed to converge to global optimum. We further propose a set of optimization techniques to speed up FXAM's training algorithm to meet the needs of interactive analysis. Evaluations verify that FXAM significantly outperforms existing GAMs in terms of training speed and modeling categorical and temporal features.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
12/15/2020

Explainable Recommendation Systems by Generalized Additive Models with Manifest and Latent Interactions

In recent years, the field of recommendation systems has attracted incre...
research
02/23/2022

Fast Sparse Classification for Generalized Linear and Additive Models

We present fast classification techniques for sparse generalized linear ...
research
02/07/2021

Additive Feature Hashing

The hashing trick is a machine learning technique used to encode categor...
research
11/07/2019

Sequence-Aware Factorization Machines for Temporal Predictive Analytics

In various web applications like targeted advertising and recommender sy...
research
08/15/2021

Exploring Temporal Coherence for More General Video Face Forgery Detection

Although current face manipulation techniques achieve impressive perform...
research
05/26/2021

SG-PALM: a Fast Physically Interpretable Tensor Graphical Model

We propose a new graphical model inference procedure, called SG-PALM, fo...
research
07/17/2014

Sparse Partially Linear Additive Models

The generalized partially linear additive model (GPLAM) is a flexible an...

Please sign up or login with your details

Forgot password? Click here to reset