Input Convex Neural Network

Input-convex neural networks (ICNNs) are a class of neural networks designed to guarantee convexity in their input-output mapping, offering advantages in training, inference, and robustness. Current research focuses on leveraging ICNNs within various optimization problems, including those in power systems, energy management, and optimal transport, often integrating them into larger frameworks like conformal prediction or deep unfolding methods. This focus stems from ICNNs' ability to provide computationally efficient and provably reliable solutions for complex, often non-convex, problems, leading to significant improvements in speed and accuracy across diverse applications.

Papers

October 16, 2024

A Primal-dual algorithm for image reconstruction with ICNNs
Hok Shing Wong, Matthias J. Ehrhardt, Subhadip Mukherjee
Image Reconstruction Variational Method Gradient Method Primal Dual Algorithm Non Smooth Input Convex Neural Network

October 1, 2024

Fast and Reliable $N-k$ Contingency Screening with Input-Convex Neural Networks
Nicolas Christianson, Wenqi Cui, Steven Low, Weiwei Yang, Baosen Zhang
Power System Input Convex Neural Network Contingency Game

September 30, 2024

End-to-End Conformal Calibration for Optimization Under Uncertainty
Christopher Yeh, Nicolas Christianson, Alan Wu, Adam Wierman, Yisong Yue
Optimization Purpose High Uncertainty Anticipation Robustness Guarantee Conformal Regression Input Uncertainty Calibrated Uncertainty Input Convex Neural Network

March 4, 2024

Locally Convex Global Loss Network for Decision-Focused Learning
Haeun Jeon, Hyunglip Bae, Minsu Park, Chanyeong Kim, Woo Chang Kim
Surrogate Loss Decision Focused Learning Input Convex Neural Network

January 15, 2024

Input Convex Lipschitz RNN: A Fast and Robust Approach for Engineering Tasks
Zihao Wang, Zhe Wu
Adversarial Robustness Robust Framework Position Engineering Solar Irradiance Input Convex Neural Network

December 19, 2023

Principled Weight Initialisation for Input-Convex Neural Networks
Pieter-Jan Hoedt, Günter Klambauer
Balancing Weight Input Convex Neural Network Non Monotonic Activation Function

November 13, 2023

Real-Time Machine-Learning-Based Optimization Using Input Convex LSTM
Zihao Wang, Donghan Yu, Zhe Wu
Long Short Term Memory Input Convex Neural Network Learning Based Optimization Based Optimization Network Based

October 6, 2023

Learning Optimal Power Flow Value Functions with Input-Convex Neural Networks
Andrew Rosemberg, Mathieu Tanneau, Bruno Fanzeres, Joaquim Garcia, Pascal Van Hentenryck
Optimal Power Flow Convex Relaxation Power Flow Input Convex Neural Network Convex Approximation

February 3, 2023

Asymmetric Certified Robustness via Feature-Convex Neural Networks
Samuel Pfrommer, Brendon G. Anderson, Julien Piet, Somayeh Sojoudi
Adversarial Robustness Convex Set Certified Robustness Input Convex Neural Network Convex Neural Network Convex Machine Learning

November 24, 2022

Deep unfolding as iterative regularization for imaging inverse problems
Zhuo-Xu Cui, Qingyong Zhu, Jing Cheng, Dong Liang
Inverse Problem Deep Unfolding Regularization Method Input Convex Neural Network Unfolded Graph Neural Network Iterative Regularization

November 11, 2022

CR-LSO: Convex Neural Architecture Optimization in the Latent Space of Graph Variational Autoencoder with Input Convex Neural Networks
Xuan Rao, Bo Zhao, Xiaosong Yi, Derong Liu
Latent Space Neural Architecture Search Variational Graph Latent Space Optimization Input Convex Neural Network Convex Neural Network Continuous Latent Space

September 18, 2022

Emission-Aware Optimization of Gas Networks: Input-Convex Neural Network Approach
Vladimir Dvorkin, Samuel Chevalier, Spyros Chatzivasileiadis
Optimization Purpose Non Convex Optimization Optimization Solver Input Convex Neural Network Natural Gas Pipeline

June 28, 2022

Supervised Training of Conditional Monge Maps
Charlotte Bunne, Andreas Krause, Marco Cuturi
Optimal Transport Input Convex Neural Network Supervised Training Monge Map Gaussian Approximation

April 14, 2022

Learning Invariances with Generalised Input-Convex Neural Networks
Vitali Nesterov, Fabricio Arend Torres, Monika Nagy-Huber, Maxim Samarin, Volker Roth
Class Manifold Input Convex Neural Network Invariance Learning

January 28, 2022

Wasserstein Iterative Networks for Barycenter Estimation
Alexander Korotin, Vage Egiazarian, Lingxiao Li, Evgeny Burnaev
Generative Model Wasserstein Gradient Wasserstein Barycenter Input Convex Neural Network Entropic Barycenter

December 4, 2021

Variational Wasserstein gradient flow
Jiaojiao Fan, Qinsheng Zhang, Amirhossein Taghvaei, Yongxin Chen
Natural Gradient Objective Function Wasserstein Gradient Flow Input Convex Neural Network Variational Formulation

November 23, 2021

Input Convex Gradient Networks
Jack Richter-Powell, Jonathan Lorraine, Brandon Amos
Input Convex Neural Network Generative Flow Model