Learning with rejection
We introduce a novel framework for classification with a rejection option that consists of
simultaneously learning two functions: a classifier along with a rejection function. We …
simultaneously learning two functions: a classifier along with a rejection function. We …
Boosting with abstention
We present a new boosting algorithm for the key scenario of binary classification with
abstention where the algorithm can abstain from predicting the label of a point, at the price of …
abstention where the algorithm can abstain from predicting the label of a point, at the price of …
Classification with costly features using deep reinforcement learning
We study a classification problem where each feature can be acquired for a cost and the
goal is to optimize a trade-off between the expected classification error and the feature cost …
goal is to optimize a trade-off between the expected classification error and the feature cost …
Theory and algorithms for learning with rejection in binary classification
We introduce a novel framework for classification with a rejection option that consists of
simultaneously learning two functions: a classifier along with a rejection function. We …
simultaneously learning two functions: a classifier along with a rejection function. We …
Pruning random forests for prediction on a budget
F Nan, J Wang, V Saligrama - Advances in neural …, 2016 - proceedings.neurips.cc
We propose to prune a random forest (RF) for resource-constrained prediction. We first
construct a RF and then prune it to optimize expected feature cost & accuracy. We pose …
construct a RF and then prune it to optimize expected feature cost & accuracy. We pose …
Feature-budgeted random forest
F Nan, J Wang, V Saligrama - International conference on …, 2015 - proceedings.mlr.press
We seek decision rules for\it prediction-time cost reduction, where complete data is available
for training, but during prediction-time, each feature can only be acquired for an additional …
for training, but during prediction-time, each feature can only be acquired for an additional …
Classification with costly features as a sequential decision-making problem
This work focuses on a specific classification problem, where the information about a sample
is not readily available, but has to be acquired for a cost, and there is a per-sample budget …
is not readily available, but has to be acquired for a cost, and there is a per-sample budget …
Efficient learning by directed acyclic graph for resource constrained prediction
J Wang, K Trapeznikov… - Advances in neural …, 2015 - proceedings.neurips.cc
We study the problem of reducing test-time acquisition costs in classification systems. Our
goal is to learn decision rules that adaptively select sensors for each example as necessary …
goal is to learn decision rules that adaptively select sensors for each example as necessary …
Model selection by linear programming
J Wang, T Bolukbasi, K Trapeznikov… - Computer Vision–ECCV …, 2014 - Springer
Budget constraints arise in many computer vision problems. Computational costs limit many
automated recognition systems while crowdsourced systems are hindered by monetary …
automated recognition systems while crowdsourced systems are hindered by monetary …
Learning a diagnostic strategy on medical data with deep reinforcement learning
M Zhu, H Zhu - IEEE Access, 2021 - ieeexplore.ieee.org
In recent years, Artificial Intelligence based disease diagnosis has drawn considerable
attention both in academia and industry. In medical scenarios, a well-trained classifier can …
attention both in academia and industry. In medical scenarios, a well-trained classifier can …