Numerous mobile apps have leveraged deep learning capabilities. However, on-device models are vulnerable to attacks as they can be easily extracted from their corresponding mobile apps. Existing on-device attacking approaches only generate black-box attacks, which are far less effective and efficient than white-box strategies. This is because mobile deep learning frameworks like TFLite do not support gradient computing, which is necessary for white-box attacking algorithms. Thus, we argue that existing findings may underestimate the harmfulness of on-device attacks. To this end, we conduct a study to answer this research question: Can on-device models be directly attacked via white-box strategies? We first systematically analyze the difficulties of transforming the on-device model to its debuggable version, and propose a Reverse Engineering framework for On-device Models (REOM), which automatically reverses the compiled on-device TFLite model to the debuggable model. Specifically, REOM first transforms compiled on-device models into Open Neural Network Exchange format, then removes the non-debuggable parts, and converts them to the debuggable DL models format that allows attackers to exploit in a white-box setting. Our experimental results show that our approach is effective in achieving automated transformation among 244 TFLite models. Compared with previous attacks using surrogate models, REOM enables attackers to achieve higher attack success rates with a hundred times smaller attack perturbations. In addition, because the ONNX platform has plenty of tools for model format exchanging, the proposed method based on the ONNX platform can be adapted to other model formats. Our findings emphasize the need for developers to carefully consider their model deployment strategies, and use white-box methods to evaluate the vulnerability of on-device models.

Machine learning (ML) is moving towards edge devices. However, ML models with high computational demands and energy consumption pose challenges for ML inference in resource-constrained environments, such as the deep sea. To address these challenges, we propose a battery-free ML inference and model personalization pipeline for microcontroller units (MCUs). As an example, we performed fish image recognition in the ocean. We evaluated and compared the accuracy, runtime, power, and energy consumption of the model before and after optimization. The results demonstrate that, our pipeline can achieve 97.78% accuracy with 483.82 KB Flash, 70.32 KB RAM, 118 ms runtime, 4.83 mW power, and 0.57 mJ energy consumption on MCUs, reducing by 64.17%, 12.31%, 52.42%, 63.74%, and 82.67%, compared to the baseline. The results indicate the feasibility of battery-free ML inference on MCUs.

Long ago, I built a hand-written digit recognition web app using Flask and TensorFlow. It was my first ML project as a beginner which didn't end up dying in a notebook, so I think it's worth sharing. This is how it's gonna look: In this tutorial, we will build our digit recognition model using TensorFlow and the MNIST dataset, which contains 70,000 images of hand-written digits 0 to 9, convert it into a TFLite model, and then build the web app. We'll be using Google Colab throughout this guide, because it's the easiest way to get started. We'll use the Keras Datasets API to load our MNIST images, because it makes it extremely easy to load the data.

In this article, we are going to convert the TensorFlow model to tflite model and will use it in a real time Sign language detection app. I hope it will be helpful. Download the Android studio and SDK file to use the android app for detection. Create the virtual device or connect your phone and run the object detection application successfully. If this Article helpful then Please hit a like and subscribe to the channel to encourage us to make more videos and articles.