Luca Zanatta

Mr. Luca Zanatta


Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
28 December 2024 in Scientific Reports
Verified
Journal Article
Open access
PDF file
LZLuca Zanatta
Luca Zanatta
FBFrancesco Barchi
Francesco Barchi
SMSimone Manoni
Simone Manoni
STSilvia Tolu
Silvia Tolu
ABAndrea Bartolini
Andrea Bartolini
AAAndrea Acquaviva
Andrea Acquaviva
Spiking Neural Networks (SNNs) stand as the third generation of Artificial Neural Networks (ANNs), mirroring the functionality of the mammalian brain more closely than their predecessors. Their computational units, spiking neurons, characterized by Ordinary Differential Equations (ODEs), allow for dynamic system representation, with spikes serving as the medium for asynchronous communication among neurons. Due to their inherent ability to capture input dynamics, SNNs hold great promise for deep networks in Reinforcement Learning (RL) tasks. Deep RL (DRL), and in particular Proximal Policy Optimization (PPO) has been proven to be valuable for training robots due to the difficulty in creating comprehensive offline datasets that capture all environmental features. DRL combined with SNNs offers a compelling solution for tasks characterized by temporal complexity. In this work, we study the effectiveness of SNNs on DRL tasks leveraging a novel framework we developed for training SNNs with PPO in the Isaac Gym simulator implemented using the skrl library. Thanks to its significantly faster training speed compared to available SNN DRL tools, the framework allowed us to: (i) Perform an effective exploration of SNN configurations for DRL robotic tasks; (ii) Compare SNNs and ANNs for various network configurations such as the number of layers and neurons. Our work demonstrates that in DRL tasks the optimal SNN topology has a lower number of layers than ANN and we highlight how the state-of-art SNN architectures used in complex RL tasks, such as Ant, SNNs have difficulties fully leveraging deeper layers. Finally, we applied the best topology identified thanks to our Isaac Gym-based framework on Ant-v4 benchmark running on MuJoCo simulator, exhibiting a performance improvement by a factor of 4.4\(\times\) over the state-of-art SNN trained on the same task.

23 August 2021 in Future Internet
Verified
Journal Article
Open access
PDF file
FBFrancesco Barchi
Francesco Barchi
LZLuca Zanatta
Luca Zanatta
EPEmanuele Parisi
Emanuele Parisi
ABAlessio Burrello
Alessio Burrello
Davide BrunelliDavide Brunelli
Davide Brunelli
Andrea AcquavivaAndrea Acquaviva
Andrea Acquaviva
In this work, we present an innovative approach for damage detection of infrastructures on-edge devices, exploiting a brain-inspired algorithm. The proposed solution exploits recurrent spiking neural networks (LSNNs), which are emerging for their theoretical energy efficiency and compactness, to recognise damage conditions by processing data from low-cost accelerometers (MEMS) directly on the sensor node. We focus on designing an efficient coding of MEMS data to optimise SNN execution on a low-power microcontroller. We characterised and profiled LSNN performance and energy consumption on a hardware prototype sensor node equipped with an STM32 embedded microcontroller and a digital MEMS accelerometer. We used a hardware-in-the-loop environment with virtual sensors generating data on an SPI interface connected to the physical microcontroller to evaluate the system with a data stream from a real viaduct. We exploited this environment also to study the impact of different on-sensor encoding techniques, mimicking a bio-inspired sensor able to generate events instead of accelerations. Obtained results show that the proposed optimised embedded LSNN (eLSNN), when using a spike-based input encoding technique, achieves 54% lower execution time with respect to a naive LSNN algorithm implementation present in the state-of-the-art. The optimised eLSNN requires around 47 kCycles, which is comparable with the data transfer cost from the SPI interface. However, the spike-based encoding technique requires considerably larger input vectors to get the same classification accuracy, resulting in a longer pre-processing and sensor access time. Overall the event-based encoding techniques leads to a longer execution time (1.49×) but similar energy consumption. Moving this coding on the sensor can remove this limitation leading to an overall more energy-efficient monitoring system.

7 October 2023 in Neurocomputing
Verified
Journal Article
Open access
LZLuca Zanatta
Luca Zanatta
AMAlfio Di Mauro
Alfio Di Mauro
FBFrancesco Barchi
Francesco Barchi
ABAndrea Bartolini
Andrea Bartolini
LBLuca Benini
Luca Benini
AAAndrea Acquaviva
Andrea Acquaviva

1 August 2024 in Neural Computing and Applications
Verified
Journal Article
Open access
PDF file
FBFrancesco Barchi
Francesco Barchi
EPEmanuele Parisi
Emanuele Parisi
LZLuca Zanatta
Luca Zanatta
Andrea BartoliniAndrea Bartolini
Andrea Bartolini
Andrea AcquavivaAndrea Acquaviva
Andrea Acquaviva
In this work, we target the efficient implementation of spiking neural networks (SNNs) for low-power and low-latency applications. In particular, we propose a methodology for tuning SNN spiking activity with the objective of reducing computation cycles and energy consumption. We performed an analysis to devise key hyper-parameters, and then we show the results of tuning such parameters to obtain a low-latency and low-energy embedded LSNN (eLSNN) implementation. We demonstrate that it is possible to adapt the firing rate so that the samples belonging to the most frequent class are processed with less spikes. We implemented the eLSNN on a microcontroller-based sensor node and we evaluated its performance and energy consumption using a structural health monitoring application processing a stream of vibrations for damage detection (i.e. binary classification). We obtained a cycle count reduction of 25% and an energy reduction of 22% with respect to a baseline implementation. We also demonstrate that our methodology is applicable to a multi-class scenario, showing that we can reduce spiking activity between 68 and 85% at iso-accuracy.


7 May 2024
Verified
Conference Paper
Open access
SMSimone Manoni
Simone Manoni
PSPaul Scheffler
Paul Scheffler
AMAlfio Di Mauro
Alfio Di Mauro
LZLuca Zanatta
Luca Zanatta
Andrea AcquavivaAndrea Acquaviva
Andrea Acquaviva
Andrea BartoliniAndrea Bartolini
Andrea Bartolini
Spiking Neural Networks (SNNs) have emerged as a promising bio-inspired solution to address the need for low-latency, energy-efficient artificial intelligence systems. SNNs pose a challenge to traditional CPUs, GPUs and neural network accelerators due to their inherent sparsity, spike-based communication between neurons and complex activation functions. Many neuromorphic accelerators have been developed to handle this complex workload, but these systems are often designed solely to accelerate spiking networks, resulting in huge area costs and a lack of flexibility. We address this problem by proposing a novel mapping methodology for Convolutional SNNs (S-CNNs) on a general-purpose open-source RISC-V core equipped with Indirection Streaming Semantic Registers, a lightweight ISA extension for accelerating sparse-dense linear algebra. NARS is the first work to map S-CNNs in a classical sparse-dense algebra paradigm. Our methodology shows that it is possible to achieve speedups on S-CNNs microkernels with sparsity degrees compatible with state-of-the-art S-CNNs ranging from 4.33× to 10.23× on a dense baseline and from 1.12× to 2.66× on a optimized dense implementation.