Impact of Hyperparameter Optimization on the Accuracy of Lightweight Deep Learning Models for Real-Time Image Classification

• URL: http://arxiv.org/abs/2507.23315v1
• Date: Thu, 31 Jul 2025 07:47:30 GMT
• Title: Impact of Hyperparameter Optimization on the Accuracy of Lightweight Deep Learning Models for Real-Time Image Classification
• Authors: Vineet Kumar Rakesh, Soumya Mazumdar, Tapas Samanta, Sarbajit Pal, Amitabha Das,
• Abstract summary: This work analyzes the influence of hyper parameter adjustment on the accuracy and convergence behavior of seven efficient deep learning architectures.<n>All models are trained on the ImageNet-1K dataset under consistent training settings.<n>Results demonstrate that cosine learning rate decay and adjustable batch size may greatly boost both accuracy and convergence speed.
• Score: 0.0
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Lightweight convolutional and transformer-based models have become vital for real-time image classification in resource-constrained applications, such as embedded systems and edge devices. This work analyzes the influence of hyperparameter adjustment on the accuracy and convergence behavior of seven efficient deep learning architectures: EfficientNetV2-S, ConvNeXt-T, MobileViT v2 (XXS/XS/S), MobileNetV3-L, TinyViT-21M, and RepVGG-A2. All models are trained on the ImageNet-1K dataset under consistent training settings, with an emphasis on real-time practicality. An comprehensive ablation study is undertaken to separate the effect of critical hyperparameters, including learning rate schedules, batch sizes, input resolution, data augmentation, regularization approaches, and optimizer choice. To assess appropriateness for real-time applications, each model is assessed not only in terms of Top-1 and Top-5 classification accuracy, but also in terms of inference time, parameter count, model size, and frames-per-second (FPS) on a GPU-accelerated edge deployment simulation. Results demonstrate that cosine learning rate decay and adjustable batch size may greatly boost both accuracy and convergence speed, while keeping low latency and memory cost. Notably, RepVGG-A2 achieves over 80% Top-1 accuracy with efficient inference performance, offering a compelling balance between accuracy and deployment cost for VGG-style models. The results give practical guidance for constructing resource-efficient deep learning models appropriate for real-time image processing pipelines. All code and training logs are publicly accessible at https://github.com/VineetKumarRakesh/lcnn-opt.

Related papers

• AHDMIL: Asymmetric Hierarchical Distillation Multi-Instance Learning for Fast and Accurate Whole-Slide Image Classification [51.525891360380285]
  AHDMIL is an Asymmetric Hierarchical Distillation Multi-Instance Learning framework.<n>It eliminates irrelevant patches through a two-step training process.<n>It consistently outperforms previous state-of-the-art methods in both classification performance and inference speed.
  arXiv  Detail & Related papers  (2025-08-07T07:47:16Z)
• Enhancing Training Data Attribution with Representational Optimization [57.61977909113113]
  Training data attribution methods aim to measure how training data impacts a model's predictions.<n>We propose AirRep, a representation-based approach that closes this gap by learning task-specific and model-aligned representations explicitly for TDA.<n>AirRep introduces two key innovations: a trainable encoder tuned for attribution quality, and an attention-based pooling mechanism that enables accurate estimation of group-wise influence.
  arXiv  Detail & Related papers  (2025-05-24T05:17:53Z)
• Comparative Analysis of Lightweight Deep Learning Models for Memory-Constrained Devices [0.0]
  Five state-of-the-art architectures are benchmarked across three diverse datasets: CIFAR-10, CIFAR-100, and Tiny ImageNet.<n>The models are assessed using four key performance metrics: classification accuracy, inference time, floating-point operations (FLOPs), and model size.
  arXiv  Detail & Related papers  (2025-05-06T08:36:01Z)
• Ultra-Resolution Adaptation with Ease [62.56434979517156]
  We propose a set of key guidelines for ultra-resolution adaptation termed emphURAE.<n>We show that tuning minor components of the weight matrices outperforms widely-used low-rank adapters when synthetic data are unavailable.<n>Experiments validate that URAE achieves comparable 2K-generation performance to state-of-the-art closed-source models like FLUX1.1 [Pro] Ultra with only 3K samples and 2K iterations.
  arXiv  Detail & Related papers  (2025-03-20T16:44:43Z)
• Building Efficient Lightweight CNN Models [0.0]
  Convolutional Neural Networks (CNNs) are pivotal in image classification tasks due to their robust feature extraction capabilities.<n>This paper introduces a methodology to construct lightweight CNNs while maintaining competitive accuracy.<n>The proposed model achieved a state-of-the-art accuracy of 99% on the handwritten digit MNIST and 89% on fashion MNIST, with only 14,862 parameters and a model size of 0.17 MB.
  arXiv  Detail & Related papers  (2025-01-26T14:39:01Z)
• A hybrid framework for effective and efficient machine unlearning [12.499101994047862]
  Machine unlearning (MU) is proposed to remove the imprints of revoked samples from the already trained model parameters.<n>We present a novel hybrid strategy on top of them to achieve an overall success.
  arXiv  Detail & Related papers  (2024-12-19T03:59:26Z)
• REP: Resource-Efficient Prompting for Rehearsal-Free Continual Learning [23.92661395403251]
  Recent rehearsal-free methods, guided by prompts, excel in vision-related continual learning (CL) with drifting data but lack resource efficiency.<n>We introduce Resource-Efficient Prompting (REP), which improves the computational and memory efficiency of prompt-based rehearsal-free methods.<n>Our approach employs swift prompt selection to refine input data using a carefully provisioned model.
  arXiv  Detail & Related papers  (2024-06-07T09:17:33Z)
• SchurVINS: Schur Complement-Based Lightweight Visual Inertial Navigation System [8.017085402991189]
  We propose a novel filter-based VINS framework named SchurVINS. It could guarantee both high accuracy by building a complete residual model and low computational complexity. Experiments on EuRoC and TUM-VI datasets show that our method notably outperforms state-of-the-art (SOTA) methods in both accuracy and computational complexity.
  arXiv  Detail & Related papers  (2023-12-04T04:14:09Z)
• Incremental Online Learning Algorithms Comparison for Gesture and Visual Smart Sensors [68.8204255655161]
  This paper compares four state-of-the-art algorithms in two real applications: gesture recognition based on accelerometer data and image classification. Our results confirm these systems' reliability and the feasibility of deploying them in tiny-memory MCUs.
  arXiv  Detail & Related papers  (2022-09-01T17:05:20Z)
• Online Convolutional Re-parameterization [51.97831675242173]
  We present online convolutional re- parameterization (OREPA), a two-stage pipeline, aiming to reduce the huge training overhead by squeezing the complex training-time block into a single convolution. Compared with the state-of-the-art re-param models, OREPA is able to save the training-time memory cost by about 70% and accelerate the training speed by around 2x. We also conduct experiments on object detection and semantic segmentation and show consistent improvements on the downstream tasks.
  arXiv  Detail & Related papers  (2022-04-02T09:50:19Z)
• A Fast and Efficient Conditional Learning for Tunable Trade-Off between Accuracy and Robustness [11.35810118757863]
  Existing models that achieve state-of-the-art (SOTA) performance on both clean and adversarially-perturbed images rely on convolution operations conditioned with feature-wise linear modulation (FiLM) layers. We present a fast learnable once-for-all adversarial training (FLOAT) algorithm, which instead of the existing FiLM-based conditioning, presents a unique weight conditioned learning that requires no additional layer. In particular, we add scaled noise to the weight tensors that enables a trade-off between clean and adversarial performance.
  arXiv  Detail & Related papers  (2022-03-28T19:25:36Z)
• Towards Scale Consistent Monocular Visual Odometry by Learning from the Virtual World [83.36195426897768]
  We propose VRVO, a novel framework for retrieving the absolute scale from virtual data. We first train a scale-aware disparity network using both monocular real images and stereo virtual data. The resulting scale-consistent disparities are then integrated with a direct VO system.
  arXiv  Detail & Related papers  (2022-03-11T01:51:54Z)
• EfficientNetV2: Smaller Models and Faster Training [91.77432224225221]
  This paper introduces EfficientNetV2, a new family of convolutional networks that have faster training speed and better parameter efficiency than previous models. We use a combination of training-aware neural architecture search and scaling, to jointly optimize training speed and parameter efficiency. Our experiments show that EfficientNetV2 models train much faster than state-of-the-art models while being up to 6.8x smaller.
  arXiv  Detail & Related papers  (2021-04-01T07:08:36Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.