Tucker Tensor Regression and Neuroimaging Analysis

This paper proposes a new methodology to predict and update the residual useful lifetime of a system using a sequence of degradation images. The methodology integrates tensor linear algebra with traditional location-scale regression widely used in reliability and prognosis. To address the high dimensionality challenge, the degradation image streams are first projected to a low-dimensional tensor subspace that is able to preserve their information. Next, the projected image tensors are regressed against time-to-failure via penalized location-scale tensor regression. The coefficient tensor is then decomposed using CANDECOMP/PARAFAC (CP) and Tucker decompositions, which enables parameter estimation in a high-dimensional setting. Two optimization algorithms with a global convergence property are developed for model estimation. The effectiveness of our models is validated using a simulated dataset and infrared degradation image streams from a rotating machinery.

show abstract

“…The proof follows the proof of Lemma 1 in Li, Zhou and Li (2013). Specifically, the mode-d matricization of tensor S i andB can be expressed as :…”

Section: A Proof Of Propositionmentioning

confidence: 92%

Image-Based Prognostics Using Penalized Tensor Regression

2019

View full text Add to dashboard Cite

show abstract

“…We next introduce an important property of the Tucker decomposition mentioned in [13]. Lemma 1 (Duality Lemma): Given a tensor pair W ∈ R I 1 ×I 2 ×···×I K and X ∈ R I 1 ×I 2 ×···×I K , and their corresponding tucker decomposed core tensor W ′ and X ′ , where W = W ′ ; B(1), .…”

Section: Tensor Operations and Tucker Decompositionmentioning

confidence: 99%

Multi-Antenna Channel Interpolation via Tucker Decomposed Extreme Learning Machine

Zhang

et al. 2019

IEEE Trans. Veh. Technol.

View full text Add to dashboard Cite

Channel interpolation is an essential technique for providing high-accuracy estimation of the channel state information (CSI) for wireless systems design where the frequency-space structural correlations of multi-antenna channel are typically hidden in matrix or tensor forms. In this letter, a modified extreme learning machine (ELM) that can process tensorial data, or ELM model with tensorial inputs (TELM), is proposed to handle the channel interpolation task. The TELM inherits many good properties from ELMs. Based on the TELM, the Tucker decomposed extreme learning machine (TDELM) is proposed for further improving the performance. Furthermore, we establish a theoretical argument to measure the interpolation capability of the proposed learning machines. Experimental results verify that our proposed learning machines can achieve comparable mean squared error (MSE) performance against the traditional ELMs but with 15% shorter running time, and outperform the other methods for a 20% margin measured in MSE for channel interpolation. Index TermsChannel interpolation, extreme learning machine, tensor decomposition.

show abstract

“…In addition, Rendle et al (2011) suggested that the choice, which is made in a specific situation, is important information for the recommender system. The CARS can use this information to make predictions e.g., Tucker tensor factorization model (Li et al 2013). Context-aware applications and solutions in health care sectors have increased significantly by technological standards in recent years.…”

Section: Related Workmentioning

confidence: 99%

Machine learning and dynamic user interfaces in a context aware nurse application environment

Ham

Dirin

Laine

2016

J Ambient Intell Human Comput

View full text Add to dashboard Cite

The increasing usage of smartphones in daily life has received considerable attention in academic and industry driven research to be utilized in the health sector. There has been development of a variety of health-related smartphone applications. Currently, however, there are few to none applications based on nurses' historical or behavioral preferences. Mobile application development for the health care sector requires extensive attention to security, reliability, and accuracy. In nursing applications, the users are often required to navigate in hospital environments, select patients to support, read the patient history and set action points to assist the patient during their shift. Finally, they have to report their performance on patient related activities and other relevant information before they leave for the day. In a working day, a nurse often visits different locations such as the patient's room, different laboratories, and offices for filling reports. There is still a limited capability to access context relevant information on a smartphone with minimal recourse such as Wi-Fi triangulation. The Wi-Fi triangulation signals fluctuate significantly for indoor location positioning. Therefore, providing relevant location based services to a mobile subscriber has become challenging. This paper addresses this gap by applying machine learning and behavior analysis to anticipate the potential location of the nurse and provide the required services. The application concept was already presented at the IMCOM 2015 conference. This paper focuses on the process to ascertain a user's context, the process of analyzing and predicting user behavior, and finally, the process of displaying the information through a dynamically generated UI.

show abstract

Tucker Tensor Regression and Neuroimaging Analysis

Cited by 121 publications

References 36 publications

Image-Based Prognostics Using Penalized Tensor Regression

Image-Based Prognostics Using Penalized Tensor Regression

Multi-Antenna Channel Interpolation via Tucker Decomposed Extreme Learning Machine

Machine learning and dynamic user interfaces in a context aware nurse application environment

Contact Info

Product

Resources

About