1,720,972 research outputs found
Vrancea earthquakes: a special challenge for seismic isolation in Bucharest
No full textIn 2000, finalizing the UNESCO-IGCP Project 414 and exploiting the existing CEI Network, neo-determinsitic maps of seismic hazard (peak amplitudes of the horizontal ground motion - displacement, velocity and design ground acceleration) were published (Panza & Vaccari, 2000). Shallow seismicity has been considered as a rule, limiting the computations to epicentral distances ≤ 90 km. The hypocentral depth considered is 10 km for events with magnitude Mw<7 and 15 km for larger events. For the Vrancea intermediate-depth events spectral properties especially determined for the Romanian these earthquakes have been considered and the computations have been performed over the Romanian, Northeastern Croatian and Hungarian territory, within a circle of 350 km of radius centered on Vrancea. The hypocentral depths considered are 90 km for M< 7.4 and 150 km for larger quakes
12 mJ Per Class On-Device Online Few-Shot Class-Incremental Learning
No full textFew-Shot Class-Incremental Learning (FSCIL) enables machine learning systems to expand their inference capabilities to new classes using only a few labeled examples, without forgetting the previously learned classes. Classical backpropagation-based learning and its variants are often unsuitable for battery-powered, memory-constrained systems at the extreme edge. In this work, we introduce Online Few-Shot Class-Incremental Learning (O-FSCIL), based on a lightweight model consisting of a pre-trained and metalearned feature extractor and an expandable explicit memory storing the class prototypes. The architecture is pretrained with a novel feature orthogonality regularization and metalearned with a multi-margin loss. For learning a new class, our approach extends the explicit memory with novel class prototypes, while the remaining architecture is kept frozen. This allows learning previously unseen classes based on only a few examples with one single pass (hence online). O-FSCIL obtains an average accuracy of 68.62% on the FSCIL CIFAR100 benchmark, achieving state-of-the-art results. Tailored for ultra-low-power platforms, we implement O-FSCIL on the 60mW GAP9 microcontroller, demonstrating online learning capabilities within just 12 mJ per new class
Work In Progress: Linear Transformers for TinyML
No full textWe present the WaveFormer, a neural network architecture based on a linear attention transformer to enable long sequence inference for TinyML devices. Waveformer achieves a new state-of-the-art accuracy of 98.8 % and 99.1 % on the Google Speech V2 keyword spotting (KWS) dataset for the 12 and 35 class problems with only 130 kB of weight storage, compatible with MCU class devices. Top-1 accuracy is improved by 0.1 and 0.9 percentage points while reducing the model size and number of operations by 2.5× and 4.7× compared to the state of the art. We also propose a hardware-friendly 8-bit integer quantization algorithm for the linear attention operator, enabling efficient deployment on low-cost, ultra-low-power microcontrollers without loss of accuracy
Deterministic Approach for the Seismic Microzonation of Bucharest
No full textThe mapping of the seismic ground motion in Bucharest, due to the strong Vrancea earthquakes is carried out using a complex hybrid waveform modeling method which combines the modal summation technique, valid for laterally homogeneous anelastic media, with finite-differences technique, and optimizes the advantages of both methods. For recent earthquakes, it is possible to validate the modeling by comparing the synthetic seismograms with the records. We consider for our computations the frequency range from 0.05 to 1.0 Hz and control the synthetic signals against the accelerograms of the Magurele station, low-pass filtered with a cut-off frequency of 1.0 Hz of the 3 last major strong (Mw > 6) Vrancea earthquakes. Using the hybrid method with a double-couple seismic source approximation, scaled for the source dimensions and relatively simple regional (bedrock) and local structure models, we succeeded in reproducing the recorded ground motion in Bucharest at a satisfactory level for seismic engineering. Extending the modeling to the entire territory of the Bucharest area, we construct a new seismic microzonation map, where five different zones are identified by their characteristic response spectra
Stargate: Multimodal Sensor Fusion for Autonomous Navigation on Miniaturized UAVs
No full textAutonomously navigating robots need to perceive and interpret their surroundings. Currently, cameras are among the most used sensors due to their high resolution and frame rates at relatively low-energy consumption and cost. In recent years, cutting-edge sensors, such as miniaturized depth cameras, have demonstrated strong potential, specifically for nano-size unmanned aerial vehicles (UAVs), where low-power consumption, lightweight hardware, and low-computational demand are essential. However, cameras are limited to working under good lighting conditions, while depth cameras have a limited range. To maximize robustness, we propose to fuse a millimeter form factor 64 pixel depth sensor and a low-resolution grayscale camera. In this work, a nano-UAV learns to detect and fly through a gate with a lightweight autonomous navigation system based on two tinyML convolutional neural network models trained in simulation, running entirely onboard in 7.6 ms and with an accuracy above 91%. Field tests are based on the Crazyflie 2.1, featuring a total mass of 39 g. We demonstrate the robustness and potential of our navigation policy in multiple application scenarios, with a failure probability down to 1.2. 10-3 crash/meter, experiencing only two crashes on a cumulative flight distance of 1.7 km
Towards On-device Domain Adaptation for Noise-Robust Keyword Spotting
No full textThe accuracy of a keyword spotting model deployed on embedded devices often degrades when the system is exposed to real environments with significant noise. In this paper, we explore a methodology for tailoring a model to on-site noises through on-device domain adaptation, while accounting 14 the edge computing-associated costs. We show that accuracy improvements by up to 18% can be obtained by specialising on difficult, previously unseen noise types, on embedded devices with a power budget in the Watt range, with a storage requirement of 1.1GB. We also demonstrate an accuracy improvement of 1.43% on an ultra-low power platform consuming few-10 mW, requiring only 1.47 MB of memory kw the adaptation stage, at a one-time energy cost of 5.81J
On-Device Domain Learning for Keyword Spotting on Low-Power Extreme Edge Embedded Systems
No full textKeyword spotting accuracy degrades when neural networks are exposed to noisy environments. On-site adaptation to previously unseen noise is crucial to recovering accuracy loss, and on-device learning is required to ensure that the adaptation process happens entirely on the edge device. In this work, we propose a fully on-device domain adaptation system achieving up to 14% accuracy gains over already-robust keyword spotting models. We enable on-device learning with less than 10 kB of memory, using only 100 labeled utterances to recover 5% accuracy after adapting to the complex speech noise. We demonstrate that domain adaptation can be achieved on ultra-low-power microcontrollers with as little as 806 mJ in only 14 s on always-on, battery-operated devices
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Variations on the Author
Full text link“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
- …
