arXiv subjects

Konstantinos Balaskas

Publications and source records attributed to Konstantinos Balaskas.

16 recordsLinked to original sources

arXiv2026

Co-Design of CNN Accelerators for TinyML using Approximate Matrix Decomposition

The paradigm shift towards local and on-device inference under stringent resource constraints is represented by the tiny machine learning (TinyML) domain. The primary goal of TinyML is to integrate intelligence into tiny, low-cost devices under strict resource, energy, and latency constraints. However, the ultra-resource-constrained nature of these devices can lead to increased inference execution time, which can be detrimental in latency critical applications. At the same time, TinyML applications are often associated with sensitive data. As such, latency optimization approaches that rely on training samples are infeasible when such data is unavailable, proprietary, or sensitive, highlighting a pressing need for optimization approaches that do not require access to the training dataset and can be applied directly to pre-trained models. Replacing costly multiplications with more hardware-efficient operations, such as shifts and additions, has been proposed as an effective method for reducing inference latency. However, post-training power-of-two (Po2) approaches are scarce and, in many cases, lead to unacceptable accuracy loss. In this work, we propose a framework that applies approximate matrix decomposition to a given CNN in order to optimize hardware implementations subject to strict constraints and without any need of re-training or fine-tuning steps. The genetic algorithm-driven framework explores different matrix decompositions and resulting multiplier-less CNN accelerator designs for FPGA targets. A comprehensive evaluation of different TinyML benchmarks demonstrates our framework's efficacy in generating latency-optimized implementations that satisfy strict accuracy and resource constraints, achieving an average 33% latency improvement with an average accuracy loss of 1.3% compared to typical systolic array-based FPGA accelerators.

cs.AR↗

arXiv2025

Design and Optimization of Mixed-Kernel Mixed-Signal SVMs for Flexible Electronics

Flexible Electronics (FE) have emerged as a promising alternative to silicon-based technologies, offering on-demand low-cost fabrication, conformality, and sustainability. However, their large feature sizes severely limit integration density, imposing strict area and power constraints, thus prohibiting the realization of Machine Learning (ML) circuits, which can significantly enhance the capabilities of relevant near-sensor applications. Support Vector Machines (SVMs) offer high accuracy in such applications at relatively low computational complexity, satisfying FE technologies' constraints. Existing SVM designs rely solely on linear or Radial Basis Function (RBF) kernels, forcing a trade-off between hardware costs and accuracy. Linear kernels, implemented digitally, minimize overhead but sacrifice performance, while the more accurate RBF kernels are prohibitively large in digital, and their analog realization contains inherent functional approximation. In this work, we propose the first mixed-kernel and mixed-signal SVM design in FE, which unifies the advantages of both implementations and balances the cost/accuracy trade-off. To that end, we introduce a co-optimization approach that trains our mixed-kernel SVMs and maps binary SVM classifiers to the appropriate kernel (linear/RBF) and domain (digital/analog), aiming to maximize accuracy whilst reducing the number of costly RBF classifiers. Our designs deliver 7.7% higher accuracy than state-of-the-art single-kernel linear SVMs, and reduce area and power by 108x and 17x on average compared to digital RBF implementations.

cs.AR↗

arXiv2025

Invited Paper: Feature-to-Classifier Co-Design for Mixed-Signal Smart Flexible Wearables for Healthcare at the Extreme Edge

Flexible Electronics (FE) offer a promising alternative to rigid silicon-based hardware for wearable healthcare devices, enabling lightweight, conformable, and low-cost systems. However, their limited integration density and large feature sizes impose strict area and power constraints, making ML-based healthcare systems-integrating analog frontend, feature extraction and classifier-particularly challenging. Existing FE solutions often neglect potential system-wide solutions and focus on the classifier, overlooking the substantial hardware cost of feature extraction and Analog-to-Digital Converters (ADCs)-both major contributors to area and power consumption. In this work, we present a holistic mixed-signal feature-to-classifier co-design framework for flexible smart wearable systems. To the best of our knowledge, we design the first analog feature extractors in FE, significantly reducing feature extraction cost. We further propose an hardware-aware NAS-inspired feature selection strategy within ML training, enabling efficient, application-specific designs. Our evaluation on healthcare benchmarks shows our approach delivers highly accurate, ultra-area-efficient flexible systems-ideal for disposable, low-power wearable monitoring.

eess.SP↗

arXiv2025

Support Vector Machines Classification on Bendable RISC-V

Flexible Electronics (FE) technology offers uniquecharacteristics in electronic manufacturing, providing ultra-low-cost, lightweight, and environmentally-friendly alternatives totraditional rigid electronics. These characteristics enable a rangeof applications that were previously constrained by the costand rigidity of conventional silicon technology. Machine learning (ML) is essential for enabling autonomous, real-time intelligenceon devices with smart sensing capabilities in everyday objects. However, the large feature sizes and high power consumption ofthe devices oppose a challenge in the realization of flexible ML applications. To address the above, we propose an open-source framework for developing ML co-processors for the Bendable RISC-V core. In addition, we present a custom ML accelerator architecture for Support Vector Machine (SVM), supporting both one-vs-one (OvO) and one-vs-rest (OvR) algorithms. Our ML accelerator adopts a generic, precision-scalable design, supporting 4-, 8-, and 16-bit weight representations. Experimental results demonstrate a 21x improvement in both inference execution time and energy efficiency, on average, highlighting its potential for low-power, flexible intelligence on the edge.

cs.AR↗

arXiv2025

Exploration of Low-Power Flexible Stress Monitoring Classifiers for Conformal Wearables

Conventional stress monitoring relies on episodic, symptom-focused interventions, missing the need for continuous, accessible, and cost-efficient solutions. State-of-the-art approaches use rigid, silicon-based wearables, which, though capable of multitasking, are not optimized for lightweight, flexible wear, limiting their practicality for continuous monitoring. In contrast, flexible electronics (FE) offer flexibility and low manufacturing costs, enabling real-time stress monitoring circuits. However, implementing complex circuits like machine learning (ML) classifiers in FE is challenging due to integration and power constraints. Previous research has explored flexible biosensors and ADCs, but classifier design for stress detection remains underexplored. This work presents the first comprehensive design space exploration of low-power, flexible stress classifiers. We cover various ML classifiers, feature selection, and neural simplification algorithms, with over 1200 flexible classifiers. To optimize hardware efficiency, fully customized circuits with low-precision arithmetic are designed in each case. Our exploration provides insights into designing real-time stress classifiers that offer higher accuracy than current methods, while being low-cost, conformable, and ensuring low power and compact size.

cs.LG↗

arXiv2025

Arbitrary Precision Printed Ternary Neural Networks with Holistic Evolutionary Approximation

Printed electronics offer a promising alternative for applications beyond silicon-based systems, requiring properties like flexibility, stretchability, conformality, and ultra-low fabrication costs. Despite the large feature sizes in printed electronics, printed neural networks have attracted attention for meeting target application requirements, though realizing complex circuits remains challenging. This work bridges the gap between classification accuracy and area efficiency in printed neural networks, covering the entire processing-near-sensor system design and co-optimization from the analog-to-digital interface-a major area and power bottleneck-to the digital classifier. We propose an automated framework for designing printed Ternary Neural Networks with arbitrary input precision, utilizing multi-objective optimization and holistic approximation. Our circuits outperform existing approximate printed neural networks by 17x in area and 59x in power on average, being the first to enable printed-battery-powered operation with under 5% accuracy loss while accounting for analog-to-digital interfacing costs.

eess.SP↗

arXiv2025

Carbon-Efficient 3D DNN Acceleration: Optimizing Performance and Sustainability

As Deep Neural Networks (DNNs) continue to drive advancements in artificial intelligence, the design of hardware accelerators faces growing concerns over embodied carbon footprint due to complex fabrication processes. 3D integration improves performance but introduces sustainability challenges, making carbon-aware optimization essential. In this work, we propose a carbon-efficient design methodology for 3D DNN accelerators, leveraging approximate computing and genetic algorithm-based design space exploration to optimize Carbon Delay Product (CDP). By integrating area-efficient approximate multipliers into Multiply-Accumulate (MAC) units, our approach effectively reduces silicon area and fabrication overhead while maintaining high computational accuracy. Experimental evaluations across three technology nodes (45nm, 14nm, and 7nm) show that our method reduces embodied carbon by up to 30% with negligible accuracy drop.

cs.AR↗

arXiv2025

Computing with Printed and Flexible Electronics

Printed and flexible electronics (PFE) have emerged as the ubiquitous solution for application domains at the extreme edge, where the demands for low manufacturing and operational cost cannot be met by silicon-based computing. Built on mechanically flexible substrates, printed and flexible devices offer unparalleled advantages in terms of form factor, bio-compatibility and sustainability, making them ideal for emerging and uncharted applications, such as wearable healthcare products or fast-moving consumer goods. Their desirable attributes stem from specialized fabrication technologies, e.g., Pragmatic's FlexIC, where advancements like ultra-thin substrates and specialized printing methods expand their hardware efficiency, and enable penetration to previously unexplored application domains. In recent years, significant focus has been on machine learning (ML) circuits for resource-constrained on-sensor and near-sensor processing, both in the digital and analog domains, as they meet the requirements of target applications by PFE. Despite their advancements, challenges like reliability, device integration and efficient memory design are still prevalent in PFE, spawning several research efforts towards cross-layer optimization and co-design, whilst showing promise for advancing printed and flexible electronics to new domains.

cs.AR↗

arXiv2025

Compact Yet Highly Accurate Printed Classifiers Using Sequential Support Vector Machine Circuits

Printed Electronics (PE) technology has emerged as a promising alternative to silicon-based computing. It offers attractive properties such as on-demand ultra-low-cost fabrication, mechanical flexibility, and conformality. However, PE are governed by large feature sizes, prohibiting the realization of complex printed Machine Learning (ML) classifiers. Leveraging PE's ultra-low non-recurring engineering and fabrication costs, designers can fully customize hardware to a specific ML model and dataset, significantly reducing circuit complexity. Despite significant advancements, state-of-the-art solutions achieve area efficiency at the expense of considerable accuracy loss. Our work mitigates this by designing area- and power-efficient printed ML classifiers with little to no accuracy degradation. Specifically, we introduce the first sequential Support Vector Machine (SVM) classifiers, exploiting the hardware efficiency of bespoke control and storage units and a single Multiply-Accumulate compute engine. Our SVMs yield on average 6x lower area and 4.6% higher accuracy compared to the printed state of the art.

cs.LG↗

arXiv2025

Late Breaking Results: Leveraging Approximate Computing for Carbon-Aware DNN Accelerators

The rapid growth of Machine Learning (ML) has increased demand for DNN hardware accelerators, but their embodied carbon footprint poses significant environmental challenges. This paper leverages approximate computing to design sustainable accelerators by minimizing the Carbon Delay Product (CDP). Using gate-level pruning and precision scaling, we generate area-aware approximate multipliers and optimize the accelerator design with a genetic algorithm. Results demonstrate reduced embodied carbon while meeting performance and accuracy requirements.

cs.AR↗

arXiv2025

Late Breaking Results: Energy-Efficient Printed Machine Learning Classifiers with Sequential SVMs

Printed Electronics (PE) provide a mechanically flexible and cost-effective solution for machine learning (ML) circuits, compared to silicon-based technologies. However, due to large feature sizes, printed classifiers are limited by high power, area, and energy overheads, which restricts the realization of battery-powered systems. In this work, we design sequential printed bespoke Support Vector Machine (SVM) circuits that adhere to the power constraints of existing printed batteries while minimizing energy consumption, thereby boosting battery life. Our results show 6.5x energy savings while maintaining higher accuracy compared to the state of the art.

cs.LG↗

arXiv2024

Energy-Aware Heterogeneous Federated Learning via Approximate DNN Accelerators

In Federated Learning (FL), devices that participate in the training usually have heterogeneous resources, i.e., energy availability. In current deployments of FL, devices that do not fulfill certain hardware requirements are often dropped from the collaborative training. However, dropping devices in FL can degrade training accuracy and introduce bias or unfairness. Several works have tackled this problem on an algorithm level, e.g., by letting constrained devices train a subset of the server neural network (NN) model. However, it has been observed that these techniques are not effective w.r.t. accuracy. Importantly, they make simplistic assumptions about devices' resources via indirect metrics such as multiply accumulate (MAC) operations or peak memory requirements. We observe that memory access costs (that are currently not considered in simplistic metrics) have a significant impact on the energy consumption. In this work, for the first time, we consider on-device accelerator design for FL with heterogeneous devices. We utilize compressed arithmetic formats and approximate computing, targeting to satisfy limited energy budgets. Using a hardware-aware energy model, we observe that, contrary to the state of the art's moderate energy reduction, our technique allows for lowering the energy requirements (by 4x) while maintaining higher accuracy.

cs.AR↗

arXiv2023

Variability-Aware Approximate Circuit Synthesis via Genetic Optimization

One of the major barriers that CMOS devices face at nanometer scale is increasing parameter variation due to manufacturing imperfections. Process variations severely inhibit the reliable operation of circuits, as the operational frequency at the nominal process corner is insufficient to suppress timing violations across the entire variability spectrum. To avoid variability-induced timing errors, previous efforts impose pessimistic and performance-degrading timing guardbands atop the operating frequency. In this work, we employ approximate computing principles and propose a circuit-agnostic automated framework for generating variability-aware approximate circuits that eliminate process-induced timing guardbands. Variability effects are accurately portrayed with the creation of variation-aware standard cell libraries, fully compatible with standard EDA tools. The underlying transistors are fully calibrated against industrial measurements from Intel 14nm FinFET in which both electrical characteristics of transistors and variability effects are accurately captured. In this work, we explore the design space of approximate variability-aware designs to automatically generate circuits of reduced variability and increased performance without the need for timing guardbands. Experimental results show that by introducing negligible functional error of merely $5.3\times 10^{-3}$, our variability-aware approximate circuits can be reliably operated under process variations without sacrificing the application performance.

cs.AR↗

arXiv2023

Hardware-Aware DNN Compression via Diverse Pruning and Mixed-Precision Quantization

Deep Neural Networks (DNNs) have shown significant advantages in a wide variety of domains. However, DNNs are becoming computationally intensive and energy hungry at an exponential pace, while at the same time, there is a vast demand for running sophisticated DNN-based services on resource constrained embedded devices. In this paper, we target energy-efficient inference on embedded DNN accelerators. To that end, we propose an automated framework to compress DNNs in a hardware-aware manner by jointly employing pruning and quantization. We explore, for the first time, per-layer fine- and coarse-grained pruning, in the same DNN architecture, in addition to low bit-width mixed-precision quantization for weights and activations. Reinforcement Learning (RL) is used to explore the associated design space and identify the pruning-quantization configuration so that the energy consumption is minimized whilst the prediction accuracy loss is retained at acceptable levels. Using our novel composite RL agent we are able to extract energy-efficient solutions without requiring retraining and/or fine tuning. Our extensive experimental evaluation over widely used DNNs and the CIFAR-10/100 and ImageNet datasets demonstrates that our framework achieves $39\%$ average energy reduction for $1.7\%$ average accuracy loss and outperforms significantly the state-of-the-art approaches.

cs.LG↗

arXiv2022

Approximate Decision Trees For Machine Learning Classification on Tiny Printed Circuits

Although Printed Electronics (PE) cannot compete with silicon-based systems in conventional evaluation metrics, e.g., integration density, area and performance, PE offers attractive properties such as on-demand ultra-low-cost fabrication, flexibility and non-toxicity. As a result, it targets application domains that are untouchable by lithography-based silicon electronics and thus have not yet seen much proliferation of computing. However, despite the attractive characteristics of PE, the large feature sizes in PE prohibit the realization of complex printed circuits, such as Machine Learning (ML) classifiers. In this work, we exploit the hardware-friendly nature of Decision Trees for machine learning classification and leverage the hardware-efficiency of the approximate design in order to generate approximate ML classifiers that are suitable for tiny, ultra-resource constrained, and battery-powered printed applications.

cs.AR↗

arXiv2022

Automated Design Approximation to Overcome Circuit Aging

Transistor aging phenomena manifest themselves as degradations in the main electrical characteristics of transistors. Over time, they result in a significant increase of cell propagation delay, leading to errors due to timing violations, since the operating frequency becomes unsustainable as the circuit ages. Conventional techniques employ timing guardbands to mitigate aging-induced delay increase, which leads to considerable performance losses from the beginning of the circuit's lifetime. Leveraging the inherent error resilience of a vast number of application domains, approximate computing was recently introduced as an aging mitigation mechanism. In this work, we present the first automated framework for generating aging-aware approximate circuits. Our framework, by applying directed gate-level netlist approximation, induces a small functional error and recovers the delay degradation due to aging. As a result, our optimized circuits eliminate aging-induced timing errors. Experimental evaluation over a variety of arithmetic circuits and image processing benchmarks demonstrates that for an average error of merely $5\times10^{-3}$, our framework completely eliminates aging-induced timing guardbands. Compared to the respective baseline circuits without timing guardbands (i.e., iso-performance evaluation), the error of the circuits generated by our framework is $1208$x smaller.

cs.AR↗