paper-with-me

홈 › Papers

Knowledge Distillation for Reservoir-based Classifier: Human Activity Recognition

2025-05-29 · Masaharu Kagiyama, Tsuyoshi Okita

This paper aims to develop an energy-efficient classifier for time-series data by introducing PatchEchoClassifier, a novel model that leverages a reservoir-based mechanism known as the Echo State Network (ESN). The model is designed for human activity recognition (HAR) using one-dimensional sensor signals and incorporates a tokenizer to extract patch-level representations. To train the model efficiently, we propose a knowledge distillation framework that transfers knowledge from a high-capacity MLP-Mixer teacher to the lightweight reservoir-based student model. Experimental evaluations on multiple HAR datasets demonstrate that our model achieves over 80 percent accuracy while significantly reducing computational cost. Notably, PatchEchoClassifier requires only about one-sixth of the floating point operations (FLOPS) compared to DeepConvLSTM, a widely used convolutional baseline. These results suggest that PatchEchoClassifier is a promising solution for real-time and energy-efficient human activity recognition in edge computing environments.

📄 PDF Abstract BibTeX arXiv:2505.22985

Code (0)

등록된 구현이 없습니다.

Tasks

Activity RecognitionEdge-computingHuman Activity RecognitionKnowledge Distillation

Methods 이 논문이 사용한 방법론

Refunds@Expedia|||How do I get a full refund from Expedia? “How do I get a full refund from Expedia? How do I get a full refund from Expedia? – Call ☎️ +1-(888) 829 (0881) or +1-805-330-4056 or +1-805-330-4056 for Quick Help &…
Average Pooling 설명 없음
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
Residual Connection 설명 없음
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Global Average Pooling Global Average Pooling is a pooling operation designed to replace fully connected layers in classical CNNs. The idea is to generate one feature map for each corresponding…
MLP-Mixer 설명 없음

Similar Papers 제목 키워드 기반

From Neural Activity to Computation: Biological Reservoirs for Pattern Recognition in Digit Classification

2025-10-07 · Ludovico Iannello, Luca Ciampi, Fabrizio Tonelli, Gabriele Lagani 외 arxiv

In this paper, we present a biologically grounded approach to reservoir computing (RC), in which a network of cultured biological neurons serves as the reservoir substrate. This system, referred to as biological reservoi…

Self-Evolutionary Reservoir Computer Based on Kuramoto Model

2023-01-25 · Zhihao Zuo, Zhongxue Gan, Yuchuan Fan, Vjaceslavs Bobrovs 외

The human brain's synapses have remarkable activity-dependent plasticity, where the connectivity patterns of neurons change dramatically, relying on neuronal activities. As a biologically inspired neural network, reservo…

model

A Light-weight Deep Human Activity Recognition Algorithm Using Multi-knowledge Distillation

2021-07-06 · Runze Chen, Haiyong Luo, Fang Zhao, Xuechun Meng 외

Inertial sensor-based human activity recognition (HAR) is the base of many human-centered mobile applications. Deep learning-based fine-grained HAR models enable accurate classification in various complex application sce…

Activity RecognitionClassificationHuman Activity RecognitionKnowledge Distillation+1

From Neurons to Computation: Biological Reservoir Computing for Pattern Recognition

2025-05-06 · Ludovico Iannello, Luca Ciampi, Gabriele Lagani, Fabrizio Tonelli 외

In this paper, we introduce a novel paradigm for reservoir computing (RC) that leverages a pool of cultured biological neurons as the reservoir substrate, creating a biological reservoir computing (BRC). This system oper…

TSAK: Two-Stage Semantic-Aware Knowledge Distillation for Efficient Wearable Modality and Model Optimization in Manufacturing Lines

2024-08-26 · Hymalai Bello, Daniel Geißler, Sungho Suh, Bo Zhou 외

Smaller machine learning models, with less complex architectures and sensor inputs, can benefit wearable sensor-based human activity recognition (HAR) systems in many ways, from complexity and cost to battery life. In th…

Activity RecognitionHuman Activity RecognitionKnowledge DistillationModel Optimization