A5046908235- Advanced Memory and Neural Computing
- Ferroelectric and Negative Capacitance Devices
- CCD and CMOS Imaging Sensors
- 2D Materials and Applications
- Neural Networks and Reservoir Computing
- Graphene research and applications
- Transition Metal Oxide Nanomaterials
- Neuroscience and Neural Engineering
- Advancements in Semiconductor Devices and Circuit Design
- Physical Unclonable Functions (PUFs) and Hardware Security
- Photoreceptor and optogenetics research
- Gas Sensing Nanomaterials and Sensors
- Electronic and Structural Properties of Oxides
- Integrated Circuits and Semiconductor Failure Analysis
- Ferroelectric and Piezoelectric Materials
- Perovskite Materials and Applications
- ZnO doping and properties
- Nanowire Synthesis and Applications
- Semiconductor Quantum Structures and Devices
- Visual Attention and Saliency Detection
- Machine Learning and ELM
- Neural dynamics and brain function
- Machine Learning in Materials Science
- Semiconductor materials and devices
Pennsylvania State University
2019-2024
Abstract Spiking neural networks (SNNs) promise to bridge the gap between artificial (ANNs) and biological (BNNs) by exploiting biologically plausible neurons that offer faster inference, lower energy expenditure, event-driven information processing capabilities. However, implementation of SNNs in future neuromorphic hardware requires encoders analogous sensory neurons, which convert external/internal stimulus into spike trains based on specific algorithm along with inherent stochasticity....
The recent decline in energy, size and complexity scaling of traditional von Neumann architecture has resurrected considerable interest brain-inspired computing. Artificial neural networks (ANNs) based on emerging devices, such as memristors, achieve brain-like computing but lack energy-efficiency. Furthermore, slow learning, incremental adaptation, false convergence are unresolved challenges for ANNs. In this article we, therefore, introduce Gaussian synapses heterostructures atomically...
The representation of external stimuli in the form action potentials or spikes constitutes basis energy efficient neural computation that emerging spiking networks (SNNs) aspire to imitate. With recent evidence suggesting information brain is more often represented by explicit firing times neurons rather than mean rates, it imperative develop novel hardware can accelerate sparse and spike-timing-based encoding. Here a medium-scale integrated circuit composed two cascaded three-stage...
Two-dimensional (2D) semiconductors possess promise for the development of field-effect transistors (FETs) at ultimate scaling limit due to their strong gate electrostatics. However, proper FET requires reduction both channel length (LCH) and contact (LC), latter which has remained a challenge increased current crowding nanoscale. Here, we investigate Au contacts monolayer MoS2 FETs with LCH down 100 nm LC 20 evaluate impact on performance. are found display ∼2.5× in ON-current, from 519 206...
Natural intelligence has many dimensions, with some of its most important manifestations being tied to learning about the environment and making behavioral changes. In primates, vision plays a critical role in learning. The underlying biological neural networks contain specialized neurons synapses which not only sense process visual stimuli but also learn adapt remarkable energy efficiency. Forgetting an active Mimicking adaptive neurobiological mechanisms for seeing, learning, forgetting...
In spite of recent advancements in artificial neural networks (ANNs), the energy efficiency, multifunctionality, adaptability, and integrated nature biological remain largely unimitated by hardware neuromorphic computing systems. Here, we exploit optoelectronic, computing, programmable memory devices based on emerging two-dimensional (2D) layered materials such as MoS2 to demonstrate a monolithically integrated, multipixel, "all-in-one" bioinspired network (BNN) capable sensing, encoding,...
Physically unclonable functions (PUFs) are an integral part of modern-day hardware security. Various types PUFs already exist, including optical, electronic, and magnetic PUFs. Here, we introduce a novel straintronic PUF (SPUF) by exploiting strain-induced reversible cracking in the contact microstructures graphene field-effect transistors (GFETs). We found that strain cycling GFETs with piezoelectric gate stack high-tensile-strength metal contacts can lead to abrupt transition some GFET...
While defects are undesirable for the reliability of electronic devices, particularly in scaled microelectronics, they have proven beneficial numerous quantum and energy-harvesting applications. However, their potential new computational paradigms, such as neuromorphic brain-inspired computing, remains largely untapped. In this study, we harness aggressively field-effect transistors based on two-dimensional semiconductors to accelerate a stochastic inference engine that offers remarkable...
Incipient ferroelectricity bridges traditional dielectrics and true ferroelectrics, enabling advanced electronic memory devices. Firstly, we report incipient in freestanding SrTiO
Abstract Doping plays a critical role in tailoring the characteristics of semiconducting materials and electronic devices. Specifically, context field-effect transistors (FETs), degenerate doping silicon channel beneath source drain regions has become essential for achieving high-performance n- p-type devices, as well significantly reducing contact resistance (R_C). In contrast, two-dimensional (2D) semiconductors have mainly relied on metal work-function engineering to lower R_C. While this...
Abstract Defects pose a significant challenge to the reliability of electronic devices, particularly when dealing with scaled dimensions in silicon microelectronic industry. Consequently, extensive efforts have been made eliminate these defects from devices through optimization growth and fabrication processes. However, realm emerging nanomaterials, such as two-dimensional semiconductors, different scenario unfolds, where are prevalent. It is crucial comprehend how can impact device...
Abstract Natural intelligence has many dimensions, and in animals, learning about the environment making behavioral changes are some of its manifestations. In primates vision plays a critical role learning. The underlying biological neural networks contain specialized neurons synapses which not only sense process visual stimuli but also learns adapts, with remarkable energy efficiency. Forgetting an active Mimicking adaptive neurobiological mechanisms for seeing, learning, forgetting can,...
Abstract Development of low-power and smart vision sensors is critical for many emerging applications including the acceleration edge intelligence. In this article, we introduce an active pixel sensor (APS) technology with in-sensor compute capability based on atomically thin two-dimensional (2D) semiconducting material such as monolayer MoS2. The presented 2D APS uses only one programmable phototransistor (1T cell), which significantly reduces area overhead allowing to fit 900 pixels in...
Abstract In spite of recent advancements in bio-realistic artificial neural networks such as spiking (SNNs), the energy efficiency, multifunctionality, adaptability, and integrated nature biological (BNNs) largely remain unimitated hardware neuromorphic computing systems. Here we exploit optoelectronic programmable memory devices based on emerging two-dimensional (2D) layered materials MoS2 to demonstrate an “all-in-one” SNN system which is capable sensing, encoding, unsupervised learning,...
Abstract In spite of recent advancements in artificial neural networks (ANNs), the energy efficiency, multifunctionality, adaptability, and integrated nature biological largely remain unimitated hardware neuromorphic computing systems. Here we exploit optoelectronic, computing, programmable memory devices based on emerging two-dimensional (2D) layered materials such as MoS2 to demonstrate a monolithically integrated, multi-pixel, “all-in-one” bio-inspired network (BNN) capable sensing,...