Search Results for author: Hiroki Matsutani

Found 13 papers, 1 papers with code

A Cost-Efficient FPGA Implementation of Tiny Transformer Model using Neural ODE

no code implementations5 Jan 2024 Ikumi Okubo, Keisuke Sugiura, Hiroki Matsutani

To mitigate the computational complexity, recently, a hybrid approach has been proposed, which uses ResNet as a backbone architecture and replaces a part of its convolution layers with an MHSA (Multi-Head Self-Attention) mechanism.

Edge-computing Quantization

An FPGA-Based Accelerator for Graph Embedding using Sequential Training Algorithm

no code implementations23 Dec 2023 Kazuki Sunaga, Keisuke Sugiura, Hiroki Matsutani

A graph embedding is an emerging approach that can represent a graph structure with a fixed-length low-dimensional vector.

Graph Embedding

A Sequential Concept Drift Detection Method for On-Device Learning on Low-End Edge Devices

no code implementations19 Dec 2022 Takeya Yamada, Hiroki Matsutani

In this case, both the neural network retraining and the proposed concept drift detection are done only by sequential computation to reduce computation cost and memory utilization.

PICO

Federated Learning of Neural ODE Models with Different Iteration Counts

no code implementations19 Aug 2022 Yuto Hoshino, Hiroki Kawakami, Hiroki Matsutani

Furthermore, we show that our approach can reduce communication size by up to 92. 4% compared with a baseline ResNet model using CIFAR-10 dataset.

Federated Learning

Addressing Gap between Training Data and Deployed Environment by On-Device Learning

1 code implementation2 Mar 2022 Kazuki Sunaga, Masaaki Kondo, Hiroki Matsutani

This article introduces a neural network based on-device learning (ODL) approach to address this issue by retraining in deployed environments.

Anomaly Detection PICO

Accelerating Distributed Deep Reinforcement Learning by In-Network Experience Sampling

no code implementations26 Oct 2021 Masaki Furukawa, Hiroki Matsutani

In distributed reinforcement learning, Actor nodes acquire experiences by interacting with a given environment and a Learner node optimizes their DQN model.

reinforcement-learning Reinforcement Learning (RL)

A Low-Cost Neural ODE with Depthwise Separable Convolution for Edge Domain Adaptation on FPGAs

no code implementations27 Jul 2021 Hiroki Kawakami, Hirohisa Watanabe, Keisuke Sugiura, Hiroki Matsutani

It is implemented on Xilinx ZCU104 board and evaluated in terms of domain adaptation accuracy, inference speed, FPGA resource utilization, and speedup rate compared to a software counterpart.

Domain Adaptation Image Classification

An Overflow/Underflow-Free Fixed-Point Bit-Width Optimization Method for OS-ELM Digital Circuit

no code implementations17 Mar 2021 Mineto Tsukada, Hiroki Matsutani

Currently there has been increasing demand for real-time training on resource-limited IoT devices such as smart sensors, which realizes standalone online adaptation for streaming data without data transfers to remote servers.

Accelerating ODE-Based Neural Networks on Low-Cost FPGAs

no code implementations31 Dec 2020 Hirohisa Watanabe, Hiroki Matsutani

In this paper, using Euler method as an ODE solver, a part of ODENet is implemented as a dedicated logic on a low-cost FPGA (Field-Programmable Gate Array) board, such as PYNQ-Z2 board.

An FPGA Acceleration and Optimization Techniques for 2D LiDAR SLAM Algorithm

no code implementations29 May 2020 Keisuke Sugiura, Hiroki Matsutani

In this paper, we propose a resource-efficient FPGA implementation for accelerating scan matching computations, which typically cause a major bottleneck in 2D LiDAR SLAM methods.

Simultaneous Localization and Mapping

An FPGA-Based On-Device Reinforcement Learning Approach using Online Sequential Learning

no code implementations10 May 2020 Hirohisa Watanabe, Mineto Tsukada, Hiroki Matsutani

In addition, we propose a combination of L2 regularization and spectral normalization for the on-device reinforcement learning so that output values of the neural network can be fit into a certain range and the reinforcement learning becomes stable.

L2 Regularization OpenAI Gym +3

An On-Device Federated Learning Approach for Cooperative Model Update between Edge Devices

no code implementations27 Feb 2020 Rei Ito, Mineto Tsukada, Hiroki Matsutani

We extend it for an on-device federated learning so that edge devices can exchange their trained results and update their model by using those collected from the other edge devices.

Anomaly Detection Federated Learning

A Neural Network-Based On-device Learning Anomaly Detector for Edge Devices

no code implementations23 Jul 2019 Mineto Tsukada, Masaaki Kondo, Hiroki Matsutani

However, (1) the iterative optimization often requires significant efforts to follow changes in the distribution of normal data (i. e., concept drift), and (2) data transfers between edge and server impose additional latency and energy consumption.

Semi-supervised Anomaly Detection Supervised Anomaly Detection +1

Cannot find the paper you are looking for? You can Submit a new open access paper.