Tensor factorization for missing data imputation in medical questionnaires by Dauwels, Justin, Garg, Lalit, Earnest, Arul, Pang, Leong Khai

“…Measures such as normalized root mean square error, bias and variance are used to assess the performance of the proposed tensor-based methods in comparison with other widely used approaches, such as mean substitution, regression imputations and k-nearest neighbor estimation. …”
Get full text
Get full text

Multi-channel EEG compression based on matrix and tensor decompositions by Srinivasan, K., Dauwels, Justin, Reddy, M. Ramasubba, Cichocki, Andrzej

“…Compression schemes for EEG signals are developed based on matrix and tensor decomposition. Various ways to arrange EEG signals into matrices and tensors are explored, and several matrix and tensor decomposition schemes are applied, including SVD, CUR, PARAFAC, the Tucker decomposition, and recent random fiber selection approaches. …”
Get full text
Get full text

Tensor decomposition based R-dimensional matrix pencil method by Wen, Fuxi, Tay, Wee Peng

“…Furthermore, it is straightforward to extend the proposed R-D tensor MP to other MP type methods, such as R-D unitary tensor MP, R-D beamspace tensor MP.…”
Get full text
Get full text
Get full text

Multi-channel EEG compression based on 3D decompositions by Srinivasan, K., Dauwels, Justin, Reddy, M. Ramasubba, Cichocki, Andrzej

“…The multi-channel EEG is represented as a three-way tensor (or 3D volume) to exploit both spatial and temporal correlations efficiently. …”
Get full text
Get full text

DNN model theft through trojan side-channel on edge FPGA accelerator by Chandrasekar, Srivatsan, Lam, Siew-Kei, Thambipillai, Srikanthan

“…In particular, our attack targets the widely-used Versatile Tensor Accelerator (VTA). A hardware trojan is employed to track the memory transactions by monitoring the AXI interface signals of VTA’s submodules. …”
Get full text

Streamlining DNN obfuscation to defend against model stealing attacks by Sun, Yidan, Lam, Siew-Kei, Jiang, Guiyuan, He, Peilan

“…However, state-of-the-art (SOTA) DNN obfuscation is time-consuming, requires expertlevel changes in existing DNN compilers (e.g., Tensor Virtual Machine (TVM)), and often relies on prior knowledge of the attack models. …”
Get full text
Get full text

Out of distribution reasoning by weakly-supervised disentangled logic variational autoencoder by Rahiminasab, Zahra, Yuhas, Michael, Easwaran, Arvind

“…Our framework consists of three steps: partitioning data based on observed generative factors, training a VAE as a logic tensor network that satisfies disentanglement rules, and run-time OOD reasoning. …”
Get full text