More Like this

1. Hybrid macro/micro level backpropagation for training deep spiking neural networks

   Jin, Yingyezhe; Zhang, Wenrui; Li, Peng (December 2018, Advances in neural information processing systems)

   Spiking neural networks (SNNs) are positioned to enable spatio-temporal information processing and ultra-low power event-driven neuromorphic hardware. However, SNNs are yet to reach the same performances of conventional deep artificial neural networks (ANNs), a long-standing challenge due to complex dynamics and non-differentiable spike events encountered in training. The existing SNN error backpropagation (BP) methods are limited in terms of scalability, lack of proper handling of spiking discontinuities, and/or mismatch between the rate coded loss function and computed gradient. We present a hybrid macro/micro level backpropagation (HM2-BP) algorithm for training multi-layer SNNs. The temporal effects are precisely captured by the proposed spike-train level post-synaptic potential (S-PSP) at the microscopic level. The rate-coded errors are defined at the macroscopic level, computed and back-propagated across both macroscopic and microscopic levels. Different from existing BP methods, HM2-BP directly computes the gradient of the rate-coded loss function w.r.t tunable parameters. We evaluate the proposed HM2-BP algorithm by training deep fully connected and convolutional SNNs based on the static MNIST [14] and dynamic neuromorphic N-MNIST [26]. HM2-BP achieves an accuracy level of 99:49% and 98:88% for MNIST and N-MNIST, respectively, outperforming the best reported performances obtained from the existing SNN BP algorithms. Furthermore, the HM2-BP produces the highest accuracies based on SNNs for the EMNIST [3] dataset, and leads to high recognition accuracy for the 16-speaker spoken English letters of TI46 Corpus [16], a challenging spatio-temporal speech recognition benchmark for which no prior success based on SNNs was reported. It also achieves competitive performances surpassing those of conventional deep learning models when dealing with asynchronous spiking streams. 

   more » « less
2. Violence Detection using 3D Convolutional Neural Networks

   Su, J.; Her, P.; Clemens, E.; Yaz, E.; Schneider, S.; Medeiros, H. (January 2022, IEEE International Conference on Advanced Video and Signal Based Surveillance)

   Accurate detection of abnormal behavior can help improve public safety. In this work, a 3D convolutional neural network (CNN) is implemented to detect violence captured by surveillance cameras. A comprehensive study of model hyper-parameter tuning is addressed to show competitive violence detection results using a general action recognition CNN without modifying the original architecture. Experimental results on three publicly available benchmark datasets show that the proposed method outperforms other sophisticated techniques designed specifically to detect violence in videos. Our analysis further indicates that reasonable network parameter adjustments can be an effective mechanism to guide the design of computer vision models in abnormal human behavior detection. 

   more » « less
3. Using Deep Learning for Flexible and Scalable Earthquake Forecasting

   https://doi.org/10.1029/2023GL103909  

   Dascher‐Cousineau, Kelian; Shchur, Oleksandr; Brodsky, Emily_E; Günnemann, Stephan (August 2023, Geophysical Research Letters)

   Abstract Seismology is witnessing explosive growth in the diversity and scale of earthquake catalogs. A key motivation for this community effort is that more data should translate into better earthquake forecasts. Such improvements are yet to be seen. Here, we introduce the Recurrent Earthquake foreCAST (RECAST), a deep‐learning model based on recent developments in neural temporal point processes. The model enables access to a greater volume and diversity of earthquake observations, overcoming the theoretical and computational limitations of traditional approaches. We benchmark against a temporal Epidemic Type Aftershock Sequence model. Tests on synthetic data suggest that with a modest‐sized data set, RECAST accurately models earthquake‐like point processes directly from cataloged data. Tests on earthquake catalogs in Southern California indicate improved fit and forecast accuracy compared to our benchmark when the training set is sufficiently long (>10⁴events). The basic components in RECAST add flexibility and scalability for earthquake forecasting without sacrificing performance. 

   more » « less
4. The domestic violence victim as COVID crisis figure

   https://doi.org/10.1007/s11186-023-09533-4  

   Sweet, Paige L.; Glenn, Maya C.; Caponi, Jacob (October 2023, Theory and Society)

   During the early months of the COVID-19 pandemic, domestic violence came to be understood as a national emergency. In this paper, we ask how and why domestic violence was constructed as a crisis specific to the pandemic. Drawing from newspaper data, we show that the domestic violence victim came to embody the violation of gendered boundaries between “public” and “private” spheres. Representations of domestic violence centered on violence spilling over the boundaries of the home, infecting the home, or the home imploding. While theorists of crisis have focused on the central role of temporality in crisis construction, and especially the performative invocation of “new time,” we argue that crisis rhetoric often relies on anxiety about the transgression of spatial boundaries. Our spatial approach to crisis has two components. First, we argue that crisis framings often invoke the idea that seemingly distinct arenas of social life are becoming disorganized or blurred. And second, because thresholds between social spaces are coded as sacred during crisis, this spatial reordering is rendered dangerous, resulting in calls to resecure boundaries. 

   more » « less
5. Sparse Transformer Hawkes Process for Long Event Sequences

   https://doi.org/10.1007/978-3-031-43424-2_11  

   Li, Zhuoqun; Sun, Mingxuan. (September 2023, Machine Learning and Knowledge Discovery in Databases: Research Track. ECML PKDD 2023)

   Large quantities of asynchronous event sequence data such as crime records, emergence call logs, and financial transactions are becoming increasingly available from various fields. These event sequences often exhibit both long-term and short-term temporal dependencies. Variations of neural network based temporal point processes have been widely used for modeling such asynchronous event sequences. However, many current architectures including attention based point processes struggle with long event sequences due to computational inefficiency. To tackle the challenge, we propose an efficient sparse transformer Hawkes process (STHP), which has two components. For the first component, a transformer with a novel temporal sparse self-attention mechanism is applied to event sequences with arbitrary intervals, mainly focusing on short-term dependencies. For the second component, a transformer is applied to the time series of aggregated event counts, primarily targeting the extraction of long-term periodic dependencies. Both components complement each other and are fused together to model the conditional intensity function of a point process for future event forecasting. Experiments on real-world datasets show that the proposed STHP outperforms baselines and achieves significant improvement in computational efficiency without sacrificing prediction performance for long sequences. 

   more » « less