Danilo Pau

Ferheen Ayaz, Idris Zakariyya, José Cano et al.

Reducing the memory footprint of Machine Learning (ML) models, particularly Deep Neural Networks (DNNs), is essential to enable their deployment into resource-constrained tiny devices. However, a disadvantage of DNN models is their vulnerability to adversarial attacks, as they can be fooled by adding slight perturbations to the inputs. Therefore, the challenge is how to create accurate, robust, and tiny DNN models deployable on resource-constrained embedded devices. This paper reports the results of devising a tiny DNN model, robust to adversarial black and white box attacks, trained with an automatic quantizationaware training framework, i.e. QKeras, with deep quantization loss accounted in the learning loop, thereby making the designed DNNs more accurate for deployment on tiny devices. We investigated how QKeras and an adversarial robustness technique, Jacobian Regularization (JR), can provide a co-optimization strategy by exploiting the DNN topology and the per layer JR approach to produce robust yet tiny deeply quantized DNN models. As a result, a new DNN model implementing this cooptimization strategy was conceived, developed and tested on three datasets containing both images and audio inputs, as well as compared its performance with existing benchmarks against various white-box and black-box attacks. Experimental results demonstrated that on average our proposed DNN model resulted in 8.3% and 79.5% higher accuracy than MLCommons/Tiny benchmarks in the presence of white-box and black-box attacks on the CIFAR-10 image dataset and a subset of the Google Speech Commands audio dataset respectively. It was also 6.5% more accurate for black-box attacks on the SVHN image dataset.

Vinamra Sharma, Danilo Pietro Pau, José Cano

The integration of large language models (LLMs) on low-power edge devices such as Raspberry Pi, known as edge language models (ELMs), has introduced opportunities for more personalized, secure, and low-latency language intelligence that is accessible to all. However, the resource constraints inherent in edge devices and the lack of robust ethical safeguards in language models raise significant concerns about fairness, accountability, and transparency in model output generation. This paper conducts a comparative analysis of text-based bias across language model deployments on edge, cloud, and desktop environments, aiming to evaluate how deployment settings influence model fairness. Specifically, we examined an optimized Llama-2 model running on a Raspberry Pi 4; GPT 4o-mini, Gemini-1.5-flash, and Grok-beta models running on cloud servers; and Gemma2 and Mistral models running on a MacOS desktop machine. Our results demonstrate that Llama-2 running on Raspberry Pi 4 is 43.23% and 21.89% more prone to showing bias over time compared to models running on the desktop and cloud-based environments. We also propose the implementation of a feedback loop, a mechanism that iteratively adjusts model behavior based on previous outputs, where predefined constraint weights are applied layer-by-layer during inference, allowing the model to correct bias patterns, resulting in 79.28% reduction in model bias.

Idris Zakariyya, Ferheen Ayaz, Mounia Kharbouche-Harrari et al.

Reducing the memory footprint of Machine Learning (ML) models, especially Deep Neural Networks (DNNs), is imperative to facilitate their deployment on resource-constrained edge devices. However, a notable drawback of DNN models lies in their susceptibility to adversarial attacks, wherein minor input perturbations can deceive them. A primary challenge revolves around the development of accurate, resilient, and compact DNN models suitable for deployment on resource-constrained edge devices. This paper presents the outcomes of a compact DNN model that exhibits resilience against both black-box and white-box adversarial attacks. This work has achieved this resilience through training with the QKeras quantization-aware training framework. The study explores the potential of QKeras and an adversarial robustness technique, Jacobian Regularization (JR), to co-optimize the DNN architecture through per-layer JR methodology. As a result, this paper has devised a DNN model employing this co-optimization strategy based on Stochastic Ternary Quantization (STQ). Its performance was compared against existing DNN models in the face of various white-box and black-box attacks. The experimental findings revealed that, the proposed DNN model had small footprint and on average, it exhibited better performance than Quanos and DS-CNN MLCommons/TinyML (MLC/T) benchmarks when challenged with white-box and black-box attacks, respectively, on the CIFAR-10 image and Google Speech Commands audio datasets.

Martin Joel Mouk Elele, Danilo Pau, Shixin Zhuang et al.

The deployment of neural networks on resource-constrained micro-controllers has gained momentum, driving many advancements in Tiny Neural Networks. This paper introduces a tiny feed-forward neural network, TinyFC, integrated into the Field-Oriented Control (FOC) of Permanent Magnet Synchronous Motors (PMSMs). Proportional-Integral (PI) controllers are widely used in FOC for their simplicity, although their limitations in handling nonlinear dynamics hinder precision. To address this issue, a lightweight 1,400 parameters TinyFC was devised to enhance the FOC performance while fitting into the computational and memory constraints of a micro-controller. Advanced optimization techniques, including pruning, hyperparameter tuning, and quantization to 8-bit integers, were applied to reduce the model's footprint while preserving the network effectiveness. Simulation results show the proposed approach significantly reduced overshoot by up to 87.5%, with the pruned model achieving complete overshoot elimination, highlighting the potential of tiny neural networks in real-time motor control applications.

Colby Banbury, Vijay Janapa Reddi, Peter Torelli et al.

Advancements in ultra-low-power tiny machine learning (TinyML) systems promise to unlock an entirely new class of smart applications. However, continued progress is limited by the lack of a widely accepted and easily reproducible benchmark for these systems. To meet this need, we present MLPerf Tiny, the first industry-standard benchmark suite for ultra-low-power tiny machine learning systems. The benchmark suite is the collaborative effort of more than 50 organizations from industry and academia and reflects the needs of the community. MLPerf Tiny measures the accuracy, latency, and energy of machine learning inference to properly evaluate the tradeoffs between systems. Additionally, MLPerf Tiny implements a modular design that enables benchmark submitters to show the benefits of their product, regardless of where it falls on the ML deployment stack, in a fair and reproducible manner. The suite features four benchmarks: keyword spotting, visual wake words, image classification, and anomaly detection.

Giulia Crocioni, Giambattista Gruosso, Danilo Pau et al.

Nowadays, Neural Networks represent a major expectation for the realization of powerful Deep Learning algorithms, which can determine several physical systems' behaviors and operations. Computational resources required for model, training, and running are large, especially when related to the amount of data that Neural Networks typically need to generalize. The latest TinyML technologies allow integrating pre-trained models on embedded systems, allowing making computing at the edge faster, cheaper, and safer. Although these technologies originated in the consumer and industrial worlds, many sectors can greatly benefit from them, such as the automotive industry. In this paper, we present a framework for implementing Neural Network-based models on a family of automotive Microcontrollers, showing their efficiency in two case studies applied to vehicles: intrusion detection on the Controller Area Network bus and residual capacity estimation in Lithium-Ion batteries, widely used in Electric Vehicles.

Colby R. Banbury, Vijay Janapa Reddi, Max Lam et al.

Recent advancements in ultra-low-power machine learning (TinyML) hardware promises to unlock an entirely new class of smart applications. However, continued progress is limited by the lack of a widely accepted benchmark for these systems. Benchmarking allows us to measure and thereby systematically compare, evaluate, and improve the performance of systems and is therefore fundamental to a field reaching maturity. In this position paper, we present the current landscape of TinyML and discuss the challenges and direction towards developing a fair and useful hardware benchmark for TinyML workloads. Furthermore, we present our four benchmarks and discuss our selection methodology. Our viewpoints reflect the collective thoughts of the TinyMLPerf working group that is comprised of over 30 organizations.

Luca Carcano, Emanuele Plebani, Danilo Pietro Pau et al.

Echo State Networks (ESN) are a class of Recurrent Neural Networks (RNN) that has gained substantial popularity due to their effectiveness, ease of use and potential for compact hardware implementation. An ESN contains the three network layers input, reservoir and readout where the reservoir is the truly recurrent network. The input and reservoir layers of an ESN are initialized at random and never trained afterwards and the training of the ESN is applied to the readout layer only. The alternative of Recursive Neural Gas (RNG) is one of the many proposals of fully-trainable reservoirs that can be found in the literature. Although some improvements in performance have been reported with RNG, to the best of authors' knowledge, no experimental comparative results are known with benchmarks for which ESN is known to yield excellent results. This work describes an accurate model of RNG together with some extensions to the models presented in the literature and shows comparative results on three well-known and accepted datasets. The experimental results obtained show that, under specific circumstances, RNG-based reservoirs can achieve better performance.

Giacomo Parigi, Angelo Stramieri, Danilo Pau et al.

Among the many possible approaches for the parallelization of self-organizing networks, and in particular of growing self-organizing networks, perhaps the most common one is producing an optimized, parallel implementation of the standard sequential algorithms reported in the literature. In this paper we explore an alternative approach, based on a new algorithm variant specifically designed to match the features of the large-scale, fine-grained parallelism of GPUs, in which multiple input signals are processed at once. Comparative tests have been performed, using both parallel and sequential implementations of the new algorithm variant, in particular for a growing self-organizing network that reconstructs surfaces from point clouds. The experimental results show that this approach allows harnessing in a more effective way the intrinsic parallelism that the self-organizing networks algorithms seem intuitively to suggest, obtaining better performances even with networks of smaller size.