Hierarchical Inference

Hierarchical inference tackles complex tasks by decomposing them into simpler sub-problems solved at different levels of a hierarchy, improving efficiency and accuracy. Current research focuses on optimizing this process in resource-constrained environments like edge devices, employing techniques like model pruning, online meta-learning, and Bayesian graph neural networks to manage the distribution of inference across local and remote resources. This approach offers significant potential for enhancing the performance and energy efficiency of machine learning systems in various applications, from image classification to cosmological modeling.

Papers

July 10, 2024

Exploring the Boundaries of On-Device Inference: When Tiny Falls Short, Go Hierarchical
Adarsh Prasad Behera, Paulius Daubaris, Iñaki Bravo, José Gallego, Roberto Morabito, Joerg Widmer, Jaya Prakash Varma Champati
Document Boundary Device Inference Little Help Hierarchical Inference Remote Inference

May 28, 2023

Neural Sculpting: Uncovering hierarchically modular task structure in neural networks through pruning and network analysis
Shreyas Malakarjun Patil, Loizos Michael, Constantine Dovrolis
Neural Network Deep Neural Network Edge Pruning Network Analysis Hierarchical Structure Modular Neural Network 3D Neural Task Structure Hierarchical Inference

April 23, 2023

The Case for Hierarchical Deep Learning Inference at the Network Edge
Ghina Al-Atat, Andrea Fresa, Adarsh Prasad Behera, Vishnu Narayanan Moothedath, James Gross, Jaya Prakash Champati
Case Relevance Edge Device Edge AI Edge Network Deep Learning Inference Hierarchical Deep Learning Local Inference Hierarchical Inference

April 3, 2023

Online Algorithms for Hierarchical Inference in Deep Learning applications at the Edge
Vishnu Narayanan Moothedath, Jaya Prakash Champati, James Gross
Machine Learning Model Extreme Edge Edge Device Online Algorithm Deep Learning Application Tiny Machine Learning Machine Learning Inference Hierarchical Inference

January 26, 2023

Inspecting class hierarchies in classification-based metric learning models
Hyeongji Kim, Pekka Parviainen, Terje Berge, Ketil Malde
Metric Learning Class Hierarchy Hierarchical Inference

November 15, 2022

Hierarchical Inference of the Lensing Convergence from Photometric Catalogs with Bayesian Graph Neural Networks
Ji Won Park, Simon Birrer, Madison Ueland, Miles Cranmer, Adriano Agnello, Sebastian Wagner-Carena, Philip J. Marshall, Aaron Roodman, the LSST Dark Energy Science Collaboration
Graph Neural Network Bayesian Posterior Probabilistic Dependency Hierarchical Bayesian Model Hierarchical Inference Mass Convergence Event Export Import Exchange Rate Convergence

August 25, 2022

Compositional Active Inference II: Polynomial Dynamics. Approximate Inference Doctrines
Toby St. Clere Smithe
Active Inference Polynomial System Approximate Inference Compositional Approach Hierarchical Inference

December 19, 2021

MISO hierarchical inference engine satisfying the law of importation with aggregation functions
Dechao Li, Qiannan Guo
Legal Text Fuzzy Logic Fuzzy Inference Custom Fraud Hierarchical Inference

Hierarchical Inference

Papers

Exploring the Boundaries of On-Device Inference: When Tiny Falls Short, Go Hierarchical

Neural Sculpting: Uncovering hierarchically modular task structure in neural networks through pruning and network analysis

The Case for Hierarchical Deep Learning Inference at the Network Edge

Online Algorithms for Hierarchical Inference in Deep Learning applications at the Edge

Inspecting class hierarchies in classification-based metric learning models

Hierarchical Inference of the Lensing Convergence from Photometric Catalogs with Bayesian Graph Neural Networks

Compositional Active Inference II: Polynomial Dynamics. Approximate Inference Doctrines

MISO hierarchical inference engine satisfying the law of importation with aggregation functions