@inproceedings{16363, author = {{Hansmeier, Tim and Kaufmann, Paul and Platzner, Marco}}, booktitle = {{GECCO '20: Proceedings of the Genetic and Evolutionary Computation Conference Companion}}, isbn = {{978-1-4503-7127-8}}, location = {{Cancún, Mexico}}, pages = {{125--126}}, publisher = {{Association for Computing Machinery (ACM)}}, title = {{{Enabling XCSF to Cope with Dynamic Environments via an Adaptive Error Threshold}}}, doi = {{10.1145/3377929.3389968}}, year = {{2020}}, } @inproceedings{20838, author = {{Lösch, Achim and Platzner, Marco}}, booktitle = {{2020 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW)}}, isbn = {{9781728174457}}, title = {{{MigHEFT: DAG-based Scheduling of Migratable Tasks on Heterogeneous Compute Nodes}}}, doi = {{10.1109/ipdpsw50202.2020.00012}}, year = {{2020}}, } @misc{21433, abstract = {{Modern machine learning (ML) techniques continue to move into the embedded system space because traditional centralized compute resources do not suit certain application domains, for example in mobile or real-time environments. Google’s TensorFlow Lite (TFLite) framework supports this shift from cloud to edge computing and makes ML inference accessible on resource-constrained devices. While it offers the possibility to partially delegate computation to hardware accelerators, there is no such “delegate” available to utilize the promising characteristics of reconfigurable hardware. This thesis incorporates modern platform FPGAs into TFLite by implementing a modular delegate framework, which allows accelerators within the programmable logic to take over the execution of neural network layers. To facilitate the necessary hardware/software codesign, the FPGA delegate is based on the operating system for reconfigurable computing (ReconOS), whose partial reconfiguration support enables the instantiation of model-tailored accelerator architectures. In the hardware back-end, a streaming-based prototype accelerator for the MobileNet model family showcases the working order of the platform, but falls short of the desired performance. Thus, it indicates the need for further exploration of alternative accelerator designs, which the delegate could automatically synthesize to meet a model’s demands.}}, author = {{Jentzsch, Felix P.}}, title = {{{Design and Implementation of a ReconOS-based TensorFlow Lite Delegate Architecture}}}, year = {{2020}}, } @article{3585, abstract = {{Existing approaches and tools for the generation of approximate circuits often lack generality and are restricted to certain circuit types, approximation techniques, and quality assurance methods. Moreover, only few tools are publicly available. This hinders the development and evaluation of new techniques for approximating circuits and their comparison to previous approaches. In this paper, we first analyze and classify related approaches and then present CIRCA, our flexible framework for search-based approximate circuit generation. CIRCA is developed with a focus on modularity and extensibility. We present the architecture of CIRCA with its clear separation into stages and functional blocks, report on the current prototype, and show initial experiments.}}, author = {{Witschen, Linus Matthias and Wiersema, Tobias and Ghasemzadeh Mohammadi, Hassan and Awais, Muhammad and Platzner, Marco}}, issn = {{0026-2714}}, journal = {{Microelectronics Reliability}}, keywords = {{Approximate Computing, Framework, Pareto Front, Accuracy}}, pages = {{277--290}}, publisher = {{Elsevier}}, title = {{{CIRCA: Towards a Modular and Extensible Framework for Approximate Circuit Generation}}}, doi = {{10.1016/j.microrel.2019.04.003}}, volume = {{99}}, year = {{2019}}, } @unpublished{16853, abstract = {{State-of-the-art frameworks for generating approximate circuits usually rely on information gained through circuit synthesis and/or verification to explore the search space and to find an optimal solution. Throughout the process, a large number of circuits may be subject to processing, leading to considerable runtimes. In this work, we propose a search which takes error bounds and pre-computed impact factors into account to reduce the number of invoked synthesis and verification processes. In our experimental results, we achieved speed-ups of up to 76x while area savings remain comparable to the reference search method, simulated annealing.}}, author = {{Witschen, Linus Matthias and Ghasemzadeh Mohammadi, Hassan and Artmann, Matthias and Platzner, Marco}}, booktitle = {{Fourth Workshop on Approximate Computing (AxC 2019)}}, keywords = {{Approximate computing, parameter selection, search space exploration, verification, circuit synthesis}}, pages = {{2}}, title = {{{Jump Search: A Fast Technique for the Synthesis of Approximate Circuits}}}, year = {{2019}}, } @inproceedings{10577, abstract = {{State-of-the-art frameworks for generating approximate circuits automatically explore the search space in an iterative process - often greedily. Synthesis and verification processes are invoked in each iteration to evaluate the found solutions and to guide the search algorithm. As a result, a large number of approximate circuits is subjected to analysis - leading to long runtimes - but only a few approximate circuits might form an acceptable solution. In this paper, we present our Jump Search (JS) method which seeks to reduce the runtime of an approximation process by reducing the number of expensive synthesis and verification steps. To reduce the runtime, JS computes impact factors for each approximation candidate in the circuit to create a selection of approximate circuits without invoking synthesis or verification processes. We denote the selection as path from which JS determines the final solution. In our experimental results, JS achieved speed-ups of up to 57x while area savings remain comparable to the reference search method, Simulated Annealing.}}, author = {{Witschen, Linus Matthias and Ghasemzadeh Mohammadi, Hassan and Artmann, Matthias and Platzner, Marco}}, booktitle = {{Proceedings of the 2019 on Great Lakes Symposium on VLSI - GLSVLSI '19}}, isbn = {{9781450362528}}, keywords = {{Approximate computing, design automation, parameter selection, circuit synthesis}}, location = {{Tysons Corner, VA, USA}}, publisher = {{ACM}}, title = {{{Jump Search: A Fast Technique for the Synthesis of Approximate Circuits}}}, doi = {{10.1145/3299874.3317998}}, year = {{2019}}, } @article{11950, abstract = {{Advances in electromyographic (EMG) sensor technology and machine learning algorithms have led to an increased research effort into high density EMG-based pattern recognition methods for prosthesis control. With the goal set on an autonomous multi-movement prosthesis capable of performing training and classification of an amputee’s EMG signals, the focus of this paper lies in the acceleration of the embedded signal processing chain. We present two Xilinx Zynq-based architectures for accelerating two inherently different high density EMG-based control algorithms. The first hardware accelerated design achieves speed-ups of up to 4.8 over the software-only solution, allowing for a processing delay lower than the sample period of 1 ms. The second system achieved a speed-up of 5.5 over the software-only version and operates at a still satisfactory low processing delay of up to 15 ms while providing a higher reliability and robustness against electrode shift and noisy channels.}}, author = {{Boschmann, Alexander and Agne, Andreas and Thombansen, Georg and Witschen, Linus Matthias and Kraus, Florian and Platzner, Marco}}, issn = {{0743-7315}}, journal = {{Journal of Parallel and Distributed Computing}}, keywords = {{High density electromyography, FPGA acceleration, Medical signal processing, Pattern recognition, Prosthetics}}, pages = {{77--89}}, publisher = {{Elsevier}}, title = {{{Zynq-based acceleration of robust high density myoelectric signal processing}}}, doi = {{10.1016/j.jpdc.2018.07.004}}, volume = {{123}}, year = {{2019}}, } @article{12967, abstract = {{Modern Boolean satisfiability solvers can emit proofs of unsatisfiability. There is substantial interest in being able to verify such proofs and also in using them for further computations. In this paper, we present an FPGA accelerator for checking resolution proofs, a popular proof format. Our accelerator exploits parallelism at the low level by implementing the basic resolution step in hardware, and at the high level by instantiating a number of parallel modules for proof checking. Since proof checking involves highly irregular memory accesses, we employ Hybrid Memory Cube technology for accelerator memory. The results show that while the accelerator is scalable and achieves speedups for all benchmark proofs, performance improvements are currently limited by the overhead of transitioning the proof into the accelerator memory.}}, author = {{Hansmeier, Tim and Platzner, Marco and Pantho, Md Jubaer Hossain and Andrews, David}}, issn = {{1939-8018}}, journal = {{Journal of Signal Processing Systems}}, number = {{11}}, pages = {{1259 -- 1272}}, title = {{{An Accelerator for Resolution Proof Checking based on FPGA and Hybrid Memory Cube Technology}}}, doi = {{10.1007/s11265-018-1435-y}}, volume = {{91}}, year = {{2019}}, } @inproceedings{15422, author = {{Ho, Nam and Kaufmann, Paul and Platzner, Marco}}, booktitle = {{World Congress on Nature and Biologically Inspired Computing (NaBIC)}}, publisher = {{Springer}}, title = {{{Optimization of Application-specific L1 Cache Translation Functions of the LEON3 Processor}}}, year = {{2019}}, } @misc{15883, author = {{Kumar Jeyakumar, Shankar}}, title = {{{Incremental learning with Support Vector Machine on embedded platforms}}}, year = {{2019}}, }