• Corpus ID: 235125712

Error Resilient Collaborative Intelligence via Low-Rank Tensor Completion

  title={Error Resilient Collaborative Intelligence via Low-Rank Tensor Completion},
  author={Lior Bragilevsky and Ivan V. Baji'c},
In the race to bring Artificial Intelligence (AI) to the edge, collaborative intelligence has emerged as a promising way to lighten the computation load on edge devices that run applications based on Deep Neural Networks (DNNs). Typically, a deep model is split at a certain layer into edge and cloud sub-models. The deep feature tensor produced by the edge sub-model is transmitted to the cloud, where the remaining computationally intensive workload is performed by the cloud sub-model. The… 

Figures and Tables from this paper



Tensor Completion Methods for Collaborative Intelligence

Three of the studied methods are existing, generic tensor completion methods, and are adapted here to recover deep feature tensor data, while the fourth method is newly developed specifically for deep feature Tensor completion, which is an important consideration in collaborative intelligence.

CALTEC: Content-Adaptive Linear Tensor Completion For Collaborative Intelligence

This paper proposes a method called Content-Adaptive Linear Tensor Completion (CALTeC), which is fast, data-adaptive, does not require pre-training, and produces better results than existing methods for tensor data recovery in collaborative intelligence.

A Fused CP Factorization Method for Incomplete Tensors

A modified CP tensor factorization framework that fuses the <inline-formula> norm constraint, sparseness, manifold, and smooth information simultaneously, which reveals the characteristics of commonly used regularizations for tensor completion in a certain sense and gives experimental guidance concerning how to use them.

Tensor completion for estimating missing values in visual data

An algorithm to estimate missing values in tensors of visual data by laying out the theoretical foundations and building a working algorithm is proposed, which is more accurate and robust than heuristic approaches.

Deep Residual Learning for Image Recognition

This work presents a residual learning framework to ease the training of networks that are substantially deeper than those used previously, and provides comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth.

Latent Space Inpainting for Loss-Resilient Collaborative Object Detection

  • I. Bajić
  • Computer Science
    ICC 2021 - IEEE International Conference on Communications
  • 2021
This paper shows that methods for image inpainting based on partial differential equations work well for the recovery of missing features in the latent space for collaborative object detection.

Very Deep Convolutional Networks for Large-Scale Image Recognition

This work investigates the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting using an architecture with very small convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers.

Collaborative Intelligence: Challenges and Opportunities

The paper surveys the current state of the art in CI, with special emphasis on signal processing-related challenges in feature compression, error resilience, privacy, and system-level design.

ImageNet Large Scale Visual Recognition Challenge

The creation of this benchmark dataset and the advances in object recognition that have been possible as a result are described, and the state-of-the-art computer vision accuracy with human accuracy is compared.

RTP: A Transport Protocol for Real-Time Applications

RTP provides end-to-end network transport functions suitable for applications transmitting real-time data over multicast or unicast network services and is augmented by a control protocol (RTCP) to allow monitoring of the data delivery in a manner scalable to large multicast networks.