Professor Hima Lakkaraju presents some of the latest advancements in machine learning models that are inherently interpretable such as rule-based models, risk scores, generalized additive models and prototype based models.

View the full playlist: https://www.youtube.com/playlist?list=PLoROMvodv4rPh6wa6PGcHH6vMG9sEIPxL

#machinelearning

0:00 Introduction
0:06 Inherently Interpretable Models
4:41 Bayesian Rule Lists: Generative Model
8:39 Pre-mined Antecedents
11:09 Interpretable Decision Sets: Desiderata
12:16 IDS: Objective Function
15:47 IDS: Optimization Procedure
16:46 Risk Scores: Examples
18:59 Objective function to learn risk scores
21:54 Generalized Additive Models (GAMs)
23:00 Formulation and Characteristics of GAMS
30:04 Prototype Selection for Interpretable Classification
34:00 Prototype Layers in Deep Learning Models
40:01 Attention Layers in Deep Learning Models