This talk presents key findings from two PhD research projects focused on energy-efficient neural network processing through event-based vision and simplicial architectures.
The first part covers event-based vision sensors, inspired by the human retina, which capture only scene changes to reduce data and energy consumption. A custom SoC utilizing content-addressable memory (CAM) was developed to efficiently process event-driven CNNs, enabling real-time applications in robotics and autonomous systems.
The second part explores simplicial processing units, an alternative to traditional convolutional layers. The research introduced the Channel-Specific Symmetric Simplicial (ChSymSim) model, improving network flexibility and efficiency. Multiple SoCs were fabricated, featuring optimized architectures and low-power designs, enhancing AI applications for embedded systems.
These efforts contribute to advancing AI hardware with biologically inspired and computationally efficient solutions.
Biography:
Pedro Julián is an Electronic Engineer, graduated from Universidad Nacional del Sur (UNS), Bahía Blanca, Argentina, in 1994, and holds a PhD in Systems Control from the same university (1999). He has served as an Associate Professor at Johns Hopkins University, USA and as Principal Investigator at CONICET. He is currently Head of the Allegro Design Center in Bahía Blanca and an Associate Professor at UNS. He has published five books, co-edited 13 conference proceedings, co-authored four patents, and over 40 articles in top international journals.
Pedro has supervised 15 PhD and 5 master’s theses and led several national research projects funded by agencies such as the National Agency for Scientific and Technological Promotion. He has received multiple awards, including the MINCyT Bernardo Houssay Award (2009) and the Engineering Award of the National Academy of Natural, Physical and Exact Sciences (2010), among others.
He is one of the founding members of the Argentine School of Micro and Nanoelectronics (EAMTA), offering training and education in integrated circuit technology since 2007.
This talk will present a brief overview of advances in ferroelectric devices and their integration into computing systems to provide novel functionality and energy efficiency in various data intensive applications. The talk will emphasize on cross-stack design opportunities in designing stacked intelligent 3D memory systems. The talk will cover circuit and architectural features leveraging the non-volatile properties of ferroelectric devices for hardware obfuscation, accelerator designs and in-memory compute structures.
Biography:
Vijaykrishnan Narayanan is an Evan Pugh University Professor and Robert Noll A. Chair Professor of Computer Science and Engineering and Electrical Engineering at The Pennsylvania State University. He is a Fellow of ACM, IEEE, AAAS and the National Academy of Inventors. He serves as associate director of DoE 3DFeM center, thrust lead for DARPA/SRC PRISM center, associate Editor-in-Chief of IEEE Micro, the academic coordinator for the India-US Defense Acceleration Ecosystem and Associate Executive Director of AI for GeoEd foundation.
Thin dielectric films are essential components of most micro- and nanoelectronic devices, and they have played a key role in the huge development that the semiconductor industry has experienced during the last 50 years. Guaranteeing the reliability of thin dielectric films has become more challenging, in light of strong demand from the market for improved performance in electronic devices. The degradation and breakdown of thin dielectrics under normal device operation has an enormous technological importance and thus it is widely investigated in traditional dielectrics (e.g., SiO2, HfO2, and Al2O3), and it should be further investigated in novel dielectric materials that might be used in future devices (e.g., layered dielectrics). Understanding not only the physical phenomena behind dielectric breakdown but also its statistics is crucial to ensure the reliability of modern and future electronic devices, and it can also be cleverly used for other applications, such as the fabrication of new-concept resistive switching devices (e.g., nonvolatile memories and electronic synapses). Here, the fundamentals of the dielectric breakdown phenomenon in traditional and future thin dielectrics are revised. The physical phenomena that trigger the onset, structural damage, breakdown statistics, device reliability, technological implications, and perspectives are described.
The INTI Bicentennial Micro and Nanoelectronics Center (CMNB) aims to encourage the sustainable development of the national industry by promoting the insertion of Micro and Nanotechnologies, generating strategic competencies, innovative technologies and intellectual property with the participation of companies producing goods and services and the academic sector. With this objective, we develop capacities, infrastructure, technological platforms and qualified human resources to provide technical assistance, development, innovation and technology transfer services, developing the national industry through the generation of new products with added value and enabling access to new markets. In this context, and based on the different lines and groups of research and development of the Center, we work on the integration of a wide variety of technologies, such as Micro and Nanofabrication, Microfluidics, Functional Printed Electronics, Sensors and Biosensors, Functional Nanomaterials, Encapsulations, Ceramic Technologies, LTCC, PCB, Microelectronic Prototyping, IoT, Artificial Vision, AI, I4.0, among others, for the development of new products and technology transfer. The objective of this presentation is to show the current state of the work lines of the different groups of the CMNB, their capabilities and the main developments linked to the area of micro and nanotechnologies and their applications.
Biography:
Electronic Engineer, graduated from UTN-FRBA in 1990. He has been professor at UTN-FRBA since then in different positions in the area of Digital Techniques and is Associate Professor in Electronic Technology since 2009. He has worked as a professional in the Center for Research and Development in Telecommunications, Electronics and Informatics (CITEI) at the National Institute of Industrial Technology (INTI) between 1990 and 2013. He completed a specialization training in MEMS Technologies (Micro Electro Mechanical Systems) working in Design, Manufacturing and Testing of MEMS devices, at the Microsystems Group of the Interuniversity Microelectronic Center (IMEC) in Leuven, Belgium, during 2001. He was Deputy Director of the Bicentennial Micro and Nanoelectronics Center (INTI-CMNB) since its creation in 2013 and currently is the Director of the same Center and head of Micro and Nanotechnologies area of INTI since 2019. He is responsible of several R&D groups at INTI-CMNB with the mision of promoting the sustainable development of the industry by mean of the insertion and transfer of Micro and Nanotechnologies to local SMEs, giving assistance for the development of new and innovative products at national level.
In 2003, Howard Huff gave a lecture entitled “From the Lab to the Fab: Transistors to Integrated Circuits” detailing in a vivid account the history of the early days of microelectronics, a history of which he was also a protagonist. This talk is inspired by that one. It extends it in time and seeks to highlight the less visible part, the research in the laboratory, within the strong interrelationship science-technology-production.
Biography:
Adrián Faigón is a physicist, with a PhD in Microelectronics from the Hebrew University of Jerusalem. In 1987 he founded and directed for more than 30 years the Laboratory of Device Physics-Microelectronics (LFDM) at the Faculty of Engineering of the UBA. Under his supervision, the undergraduate and doctoral theses of two dozen students were developed there. Some of them now occupy positions in related areas in industry and academia. He is co-author of a hundred publications, all related to Physics involved in contemporary technological developments in MOS technology such as gate tunneling, dielectric breakdown, high-k dielectrics, and to a specific area of application: radiation sensors based on MOS structures.
He was a Full Professor at UBA and Principal Researcher at CONICET. Retired a few years ago, he collaborates with the research of the LFDM as a Consulting Professor.
The quickly growing demand for computational power required to develop and operate
advanced artificial intelligence models, fueled by an unending desire for information access
anytime and anywhere, is using up large quantities of energy, depleting Earth’s resources, and
worsening climate change rapidly. Motivated by the effectiveness, efficiency, and robustness of
natural intelligence in biological information processing systems, neuromorphic engineering
provides viable options for extremely low-energy cognitive computing in highly parallel
distributed architecture. The emergence of nanoscale memristors, combined with
neuromorphic hardware, raised hopes of being able to build CMOL (CMOS/nanowire/molecular)
type ultra-dense in-memory-computing circuit architectures by fabricating lower density
neurons on CMOS and placing massive analog synaptic connectivity with nanowire and
nanoscale-memristor fabric post-fabricated on top. Currently available hybrid CMOS-memristor
technologies allow to implement large-scale neuromorphic systems with dense memristive
crossbars. These memristive devices are especially suited to implement online learning
algorithms like Spike-Timing-Dependent Plasticity (STDP).
Biography:
Luis A. Camuñas Mesa is an Electronic Engineer from the University of Seville, Spain (2003), and holds a PhD in event-based vision systems from the same university (2010). In 2006, he was a Visiting Student with the Institute of Neuroinformatics, Zürich (Switzerland). From 2010 to 2013, he was an Associate Post-Doctoral Researcher with the Centre for Systems Neuroscience, University of Leicester (UK), where he was involved in olfactory sensing and processing, spike detection and sorting, and simulation of extracellular recordings. In 2013, he received the ‘Juan de la Cierva’ Post-Doctoral Fellowship at the Institute of Microelectronics of Seville (IMSE-CNM). In 2019, he became a Research Associate at the University of Seville. Since 2023, he holds a position as Tenured Scientist at CSIC (Spanish National Research Council). His current research interests include bioinspired circuits and systems, real-time event-based vision sensing and processing chips, neuromorphic stereo vision, and nanoscale memristor-based AER circuits for STDP learning. He has been an Associate Editor of the IEEE Transactions of Circuits and Systems – II: Express Briefs.