Showing papers by "Shahin Nazarian published in 2022"
TL;DR: The current monolithic programming models and task mapping to compute engines do not fully exploit the recent architectural innovations and can exacerbate the load imbalance and communication inefficiencies.
Abstract: The recent technological advances have significantly contributed to a rapid increase in the algorithmic complexity of various applications, from digital signal processing to autonomous aerial, ground and underwater systems [1]. In order to control and manage this increased algorithmic complexity, heterogeneous computing systems require intelligent, flexible, and highly efficient programming strategies to provide high performance while minimizing energy costs [2, 3]. However, the current monolithic programming models and task mapping to compute engines do not fully exploit the recent architectural innovations and can exacerbate the load imbalance and communication inefficiencies [4].
TL;DR: Li et al. as mentioned in this paper proposed a trust-based pooling layer for CNNs to achieve higher accuracy and trustworthiness in applications with noise in input features, and further proposed TrustCNets consisting of trust-aware CNN building blocks.
Abstract: Convolutional neural networks (CNNs) are known to be effective tools in many deep learning application areas. Despite CNN’s good performance in terms of classical evaluation metrics such as accuracy and loss, quantifying and ensuring a high degree of trustworthiness of such models remains an unsolved problem raising questions in applications where trust is an important factor. In this work, we propose a framework to evaluate the trustworthiness of CNNs. Towards this end, we develop a trust-based pooling layer for CNNs to achieve higher accuracy and trustworthiness in applications with noise in input features. We further propose TrustCNets consisting of trustworthiness-aware CNN building blocks, i.e