Search Results for author: Margo Seltzer

Found 19 papers, 15 papers with code

Optimal Sparse Survival Trees

1 code implementation27 Jan 2024 Rui Zhang, Rui Xin, Margo Seltzer, Cynthia Rudin

Interpretability is crucial for doctors, hospitals, pharmaceutical companies and biotechnology corporations to analyze and make decisions for high stakes problems that involve human health.

Survival Analysis

CAT-Walk: Inductive Hypergraph Learning via Set Walks

1 code implementation NeurIPS 2023 Ali Behrouz, Farnoosh Hashemi, Sadaf Sadeghian, Margo Seltzer

Our evaluation on 10 hypergraph benchmark datasets shows that CAT-Walk attains outstanding performance on temporal hyperedge prediction benchmarks in both inductive and transductive settings.

Hyperedge Prediction Node Classification +1

Exploring and Interacting with the Set of Good Sparse Generalized Additive Models

1 code implementation NeurIPS 2023 Chudi Zhong, Zhi Chen, Jiachang Liu, Margo Seltzer, Cynthia Rudin

In real applications, interaction between machine learning models and domain experts is critical; however, the classical machine learning paradigm that usually produces only a single model does not facilitate such interaction.

Additive models

Optimal Sparse Regression Trees

1 code implementation28 Nov 2022 Rui Zhang, Rui Xin, Margo Seltzer, Cynthia Rudin

Regression trees are one of the oldest forms of AI models, and their predictions can be made without a calculator, which makes them broadly useful, particularly for high-stakes applications.

Clustering regression

Anomaly Detection in Multiplex Dynamic Networks: from Blockchain Security to Brain Disease Prediction

1 code implementation15 Nov 2022 Ali Behrouz, Margo Seltzer

The problem of identifying anomalies in dynamic networks is a fundamental task with a wide range of applications.

Anomaly Detection Disease Prediction

Fast Optimization of Weighted Sparse Decision Trees for use in Optimal Treatment Regimes and Optimal Policy Design

no code implementations13 Oct 2022 Ali Behrouz, Mathias Lecuyer, Cynthia Rudin, Margo Seltzer

Specifically, they rely on the discreteness of the loss function, which means that real-valued weights cannot be directly used.

FasterRisk: Fast and Accurate Interpretable Risk Scores

1 code implementation12 Oct 2022 Jiachang Liu, Chudi Zhong, Boxuan Li, Margo Seltzer, Cynthia Rudin

Specifically, our approach produces a pool of almost-optimal sparse continuous solutions, each with a different support set, using a beam-search algorithm.

Exploring the Whole Rashomon Set of Sparse Decision Trees

2 code implementations16 Sep 2022 Rui Xin, Chudi Zhong, Zhi Chen, Takuya Takagi, Margo Seltzer, Cynthia Rudin

We show three applications of the Rashomon set: 1) it can be used to study variable importance for the set of almost-optimal trees (as opposed to a single tree), 2) the Rashomon set for accuracy enables enumeration of the Rashomon sets for balanced accuracy and F1-score, and 3) the Rashomon set for a full dataset can be used to produce Rashomon sets constructed with only subsets of the data set.

Fast Sparse Classification for Generalized Linear and Additive Models

2 code implementations23 Feb 2022 Jiachang Liu, Chudi Zhong, Margo Seltzer, Cynthia Rudin

For fast sparse logistic regression, our computational speed-up over other best-subset search techniques owes to linear and quadratic surrogate cuts for the logistic loss that allow us to efficiently screen features for elimination, as well as use of a priority queue that favors a more uniform exploration of features.

Additive models Classification

Fast Sparse Decision Tree Optimization via Reference Ensembles

3 code implementations1 Dec 2021 Hayden McTavish, Chudi Zhong, Reto Achermann, Ilias Karimalis, Jacques Chen, Cynthia Rudin, Margo Seltzer

We show that by using these guesses, we can reduce the run time by multiple orders of magnitude, while providing bounds on how far the resulting trees can deviate from the black box's accuracy and expressive power.

Interpretable Machine Learning

ReLUSyn: Synthesizing Stealthy Attacks for Deep Neural Network Based Cyber-Physical Systems

no code implementations21 May 2021 Aarti Kashyap, Syed Mubashir Iqbal, Karthik Pattabiraman, Margo Seltzer

These attacks, which we call Ripple False Data Injection Attacks (rfdia), use minimal input perturbations to stealthily change the dnn output.

Collision Avoidance Management

Generalized and Scalable Optimal Sparse Decision Trees

2 code implementations ICML 2020 Jimmy Lin, Chudi Zhong, Diane Hu, Cynthia Rudin, Margo Seltzer

Decision tree optimization is notoriously difficult from a computational perspective but essential for the field of interpretable machine learning.

Interpretable Machine Learning

Optimal Sparse Decision Trees

2 code implementations NeurIPS 2019 Xiyang Hu, Cynthia Rudin, Margo Seltzer

Decision tree algorithms have been among the most popular algorithms for interpretable (transparent) machine learning since the early 1980's.

Runtime Analysis of Whole-System Provenance

1 code implementation18 Aug 2018 Thomas Pasquier, Xueyuan Han, Thomas Moyer, Adam Bates, Olivier Hermant, David Eyers, Jean Bacon, Margo Seltzer

Identifying the root cause and impact of a system intrusion remains a foundational challenge in computer security.

Cryptography and Security Operating Systems

Learning Certifiably Optimal Rule Lists for Categorical Data

5 code implementations6 Apr 2017 Elaine Angelino, Nicholas Larus-Stone, Daniel Alabi, Margo Seltzer, Cynthia Rudin

We present the design and implementation of a custom discrete optimization technique for building rule lists over a categorical feature space.

Scalable Bayesian Rule Lists

7 code implementations ICML 2017 Hongyu Yang, Cynthia Rudin, Margo Seltzer

They have a logical structure that is a sequence of IF-THEN rules, identical to a decision list or one-sided decision tree.

Computational Efficiency

Accelerating MCMC via Parallel Predictive Prefetching

no code implementations28 Mar 2014 Elaine Angelino, Eddie Kohler, Amos Waterland, Margo Seltzer, Ryan P. Adams

We present a general framework for accelerating a large class of widely used Markov chain Monte Carlo (MCMC) algorithms.

Bayesian Inference

Cannot find the paper you are looking for? You can Submit a new open access paper.