Robust Multi-Modal Sensor Fusion: An Adversarial Approach

Virtual: https://events.vtools.ieee.org/m/284533

Abstract. Increasingly important Multi-Domain Operations entail multi-modal sensing which in turn, may require much effort to harness the information for exploitation. As a result, an upsurge in research interest in multi-modal fusion has emerged across industry, and in academia and government. Combining this multi-sensor information has been of particular interest in inference and target detection to enhance performance in challenging and adversarial environments. While fusion is strictly not new, a more principled approach has only recently been emerging on account of its ubiquitous need. The viability of these fusion approaches strongly hinges on the simultaneous functionality of all the sensors, limiting their efficacy in a real environment. The severity of this limitation is even more pronounced in unconstrained surveillance settings where the environmental conditions have a direct impact on the sensors, and close manual monitoring is difficult or even impractical. Partial sensor failure can hence cause a major drop in performance in a fusion system in the absence of a timely failure detection. We will describe in this talk a data driven approach to multimodal fusion, where optimal features for each sensor are selected from a hidden latent space among different modalities. This hidden space is learned via a generative network conditioned on individual sensor modalities. The hidden space, as an intrinsic structure, is then exploited as a palliative proxy for not only detecting damaged sensors, but for subsequently safeguarding the performance of the fused sensor system. Experimental results show that such an approach can make an inference system robust against noisy/damaged sensors, without requiring human intervention to inform the system about the damage. Co-sponsored by: IEEE SP Atlanta Chapter & IEEE AESS/GRSS Atlanta chapter Speaker(s): Dr. Hamid Krim, Virtual: https://events.vtools.ieee.org/m/284533

Robust Multi-Modal Sensor Fusion: An Adversarial Approach

Virtual: https://events.vtools.ieee.org/m/284533

Abstract. Increasingly important Multi-Domain Operations entail multi-modal sensing which in turn, may require much effort to harness the information for exploitation. As a result, an upsurge in research interest in multi-modal fusion has emerged across industry, and in academia and government. Combining this multi-sensor information has been of particular interest in inference and target detection to enhance performance in challenging and adversarial environments. While fusion is strictly not new, a more principled approach has only recently been emerging on account of its ubiquitous need. The viability of these fusion approaches strongly hinges on the simultaneous functionality of all the sensors, limiting their efficacy in a real environment. The severity of this limitation is even more pronounced in unconstrained surveillance settings where the environmental conditions have a direct impact on the sensors, and close manual monitoring is difficult or even impractical. Partial sensor failure can hence cause a major drop in performance in a fusion system in the absence of a timely failure detection. We will describe in this talk a data driven approach to multimodal fusion, where optimal features for each sensor are selected from a hidden latent space among different modalities. This hidden space is learned via a generative network conditioned on individual sensor modalities. The hidden space, as an intrinsic structure, is then exploited as a palliative proxy for not only detecting damaged sensors, but for subsequently safeguarding the performance of the fused sensor system. Experimental results show that such an approach can make an inference system robust against noisy/damaged sensors, without requiring human intervention to inform the system about the damage. Co-sponsored by: IEEE SP Atlanta Chapter & IEEE AESS/GRSS Atlanta chapter Speaker(s): Dr. Hamid Krim, Virtual: https://events.vtools.ieee.org/m/284533

Landscape of Synaptic Weight Memories

Virtual: https://events.vtools.ieee.org/m/284899

The Electron Devices Society (EDS), Northern Virginia/Washington Jt. Sections Chapters joined with The Nanotechology Council (NTC) are pleased to host an EDS Distinguished Lecture presented by Prof. Shimeng Yu, School of Electrical and Computer Engineering, Georgia Institute of Technology. Please register to receive the WebEx link the day before the event. ------------------------------------------------------------------------------ Analog multilevel memories are the enabling device technologies for hardware acceleration of neuro-inspired computing workloads. In this lecture, we will survey the landscape of the emerging non-volatile memories that could serve the synaptic weights with a focus on resistive and ferroelectric devices. We will highlight the key device properties that are required for on-chip inference and/or training of deep neural network (DNN) models. We will use a multi-bit RRAM test vehicle to characterize the variability/reliability at array-level for inference. Then we will introduce an end-to-end benchmark framework DNN+NeuroSim to that is interfaced with PyTorch to evaluate versatile device technologies for DNN inference. Hybrid precision synapse that combines non-volatile memories with volatile capacitor is also presented to achieve in-situ training accuracy that is comparable with software. We will also showcase the integration of RRAM with peripheral CMOS at 40nm for a complete compute-in-memory prototype chip. Future research directions will be discussed. Speaker(s): Prof. Shimeng Yu, Virtual: https://events.vtools.ieee.org/m/284899

SMART Power Flow Controllers – A Necessity for Future Power Grid

Virtual: https://events.vtools.ieee.org/m/282280

Power flow control techniques have been practiced, from using inductors, capacitors, transformers and load tap changers in the earlier days of electrical engineering to power electronics-based solutions in recent years. Even though the costs and complexities of the available solutions vary widely, the basic underlying theory of power flow control is still the same as it always has been. The question is which solution one should employ. The answer depends on knowing what the true need is. The power industry’s pressing need for the most economical ways to transfer bulk power along a desired path may be met by building new transmission lines, which is a long and costly process. Alternately, it may be quicker and cheaper to utilize the existing transmission lines more efficiently. The key is to identify the underutilized transmission lines and harness their dormant capacity to increase the power flows to the lines’ thermal limits. The presentation is designed to provide the basic principles of power flow control theory, an overview of the most commonly used power flow controllers, and future trends. The audience will hear from an expert who actually designed and commissioned a number of power electronics-based FACTS controllers since their inception in the 1990s Speaker(s): Kalyan K. Sen, PhD, PE, MBA, IEEE Fellow, Virtual: https://events.vtools.ieee.org/m/282280

SMART Power Flow Controllers – A Necessity for Future Power Grid

Virtual: https://events.vtools.ieee.org/m/282280

Power flow control techniques have been practiced, from using inductors, capacitors, transformers and load tap changers in the earlier days of electrical engineering to power electronics-based solutions in recent years. Even though the costs and complexities of the available solutions vary widely, the basic underlying theory of power flow control is still the same as it always has been. The question is which solution one should employ. The answer depends on knowing what the true need is. The power industry’s pressing need for the most economical ways to transfer bulk power along a desired path may be met by building new transmission lines, which is a long and costly process. Alternately, it may be quicker and cheaper to utilize the existing transmission lines more efficiently. The key is to identify the underutilized transmission lines and harness their dormant capacity to increase the power flows to the lines’ thermal limits. The presentation is designed to provide the basic principles of power flow control theory, an overview of the most commonly used power flow controllers, and future trends. The audience will hear from an expert who actually designed and commissioned a number of power electronics-based FACTS controllers since their inception in the 1990s Speaker(s): Kalyan K. Sen, PhD, PE, MBA, IEEE Fellow, Virtual: https://events.vtools.ieee.org/m/282280

IEEE Photonics Society Distinguished Lecture: Nitrides for quantum light sources

Virtual: https://events.vtools.ieee.org/m/285280

Speaker: Prof. Rachel Oliver, Department of Materials Science and Metallurgy, University of Cambridge Abstract: A quantum light source is a device that can generate one single photon – or an entangled pair of photons - on demand. Whilst a single photon emitter would be pretty useless as a car headlight or bedside lamp, these devices are in increasing demand for new developments in optical communication which might exploit fundamental principles of quantum physics to achieve data security. Linear optical quantum computation, precision optical measurement and even random number generation also present potential applications opportunities for such light sources. However, many of the most mature quantum light sources operate at temperatures only accessible using liquid helium, at best inconvenient and at worst prohibitive for applications. Exploiting nitride semiconductors allows device concepts developed in the more conventional arsenide semiconductor family to be applied, but whilst arsenide devices are limited to cryogenic temperatures, nitride devices can operate at temperatures accessible using on-chip, Peltier cooling, and in some cases even at room temperature. Unfortunately, working with these less mature semiconductors has its pitfalls: high densities of defects and the impact of internal electric fields can limit device performance. For example, the wavelength of emission from nitride single photon emitters wanders with time, which is not compatible with applications which demand resonance of the emitter with a cavity or (more stringently) the emission of indistinguishable photons. Nitrides crystals grown in unusual orientations can overcome these challenges whilst maintaining good temperature stability, providing new opportunities for real-world quantum technologies. Bio: Professor Rachel Oliver is Director of the Cambridge Centre for Gallium Nitride. She leads research projects across the full range of the Centre’s activities, and her personal passion is understanding how the small scale structure of nitride materials effects the performance and properties of devices. She uses expertise in microscopy and materials growth to lead the development of new nanoscale nitride structures which will provide new functionality to the devices of the future. She is also passionate about communicating science to the general public and hence widening participation in science by under-represented groups, particularly women. Dr. Oliver is a Fellow of the Royal Academy of Engineering. Virtual: https://events.vtools.ieee.org/m/285280

IEEE Photonics Society Distinguished Lecture: Nitrides for quantum light sources

Virtual: https://events.vtools.ieee.org/m/285280

Speaker: Prof. Rachel Oliver, Department of Materials Science and Metallurgy, University of Cambridge Abstract: A quantum light source is a device that can generate one single photon – or an entangled pair of photons - on demand. Whilst a single photon emitter would be pretty useless as a car headlight or bedside lamp, these devices are in increasing demand for new developments in optical communication which might exploit fundamental principles of quantum physics to achieve data security. Linear optical quantum computation, precision optical measurement and even random number generation also present potential applications opportunities for such light sources. However, many of the most mature quantum light sources operate at temperatures only accessible using liquid helium, at best inconvenient and at worst prohibitive for applications. Exploiting nitride semiconductors allows device concepts developed in the more conventional arsenide semiconductor family to be applied, but whilst arsenide devices are limited to cryogenic temperatures, nitride devices can operate at temperatures accessible using on-chip, Peltier cooling, and in some cases even at room temperature. Unfortunately, working with these less mature semiconductors has its pitfalls: high densities of defects and the impact of internal electric fields can limit device performance. For example, the wavelength of emission from nitride single photon emitters wanders with time, which is not compatible with applications which demand resonance of the emitter with a cavity or (more stringently) the emission of indistinguishable photons. Nitrides crystals grown in unusual orientations can overcome these challenges whilst maintaining good temperature stability, providing new opportunities for real-world quantum technologies. Bio: Professor Rachel Oliver is Director of the Cambridge Centre for Gallium Nitride. She leads research projects across the full range of the Centre’s activities, and her personal passion is understanding how the small scale structure of nitride materials effects the performance and properties of devices. She uses expertise in microscopy and materials growth to lead the development of new nanoscale nitride structures which will provide new functionality to the devices of the future. She is also passionate about communicating science to the general public and hence widening participation in science by under-represented groups, particularly women. Dr. Oliver is a Fellow of the Royal Academy of Engineering. Virtual: https://events.vtools.ieee.org/m/285280

IEEE IAS Atlanta In-Person Meeting (High Resistance Grounding)

Critical Components, Inc., 120 Interstate North Pkwy SE, Suite 305, Atlanta, Georgia, United States, 30339, Virtual: https://events.vtools.ieee.org/m/286009

You are invited to join us for lunch and learning at the IEEE Industry Applications Society - Atlanta Chapter meeting. In-person Meeting Cost: Free (Limited to 15 people with valid IEEE membership) Speaker(s): Joe Campa, Agenda: Registration: 11:30 AM Meeting Starts: 12:00 Noon Critical Components, Inc., 120 Interstate North Pkwy SE, Suite 305, Atlanta, Georgia, United States, 30339, Virtual: https://events.vtools.ieee.org/m/286009