You have full access to this article via your institution.
…
Category: 7. Maths
-
DelGrad: exact event-based gradients for training delays and weights on spiking neuromorphic hardware
Training delays in SNNs with DelGrad
While spiking neural networks share their overall structure with the more well-known artificial neural networks, the intrinsic dynamics in the networks are different. In SNNs, each unit in the network is a…
Continue Reading
-
Analog in-memory computing attention mechanism for fast and energy-efficient large language models
Hardware-based neural network simulations
We implement the sliding window attention by masking the elements of S outside the sliding window (blank spaces in the example Fig. 1). The HardSigmoid charge-to-pulse circuit is modeled by the equation
Continue Reading
-
Urban planning in the era of large language models
World Cities Report 2020: The Value of Sustainable Urbanization (UN-Habitat, 2020).
Su, H., Zheng, Y., Ding, J., Jin, D. & Li, Y. Large-scale urban facility location selection with knowledge-informed reinforcement learning. In Proc. 32nd ACM…
Continue Reading
-
Overcoming computational bottlenecks in large language models through analog in-memory computing
de Vries, A. Joule 7, 2191–2194 (2023).
Google Scholar
Leroux, N. et al. Nat. Comput. Sci. https://doi.org/10.1038/s43588-025-00854-1 (2025).
Google…
Continue Reading
-
Towards compute-efficient Byzantine-robust federated learning with fully homomorphic encryption
Accountability Act Health insurance portability and accountability act of 1996. Public Law 104, 191 (1996).
European Parliament & Council. General data protection regulation. Off. J. Eur. Union…
Continue Reading
-
Three methods for fair ranking of multiple protected items
The experimental study was designed to quantify the extent to which a state-of-the-art system can rank the retrieval results fairly and effectively after using a post-processing technique. The code is available at31.
The test collections from the…
Continue Reading
-
Covariant spatio-temporal receptive fields for spiking neural networks
Covariance properties of spatio-temporal receptive fields
Consider a composition of the spatio-temporal image transformations defined in Section Joint covariance under geometric image transformations of the form
$${x}^{{\prime} }=A\,(x+u\,t),$$
…
Continue Reading
-
Dual-self-learning co-evolutionary algorithm for energy-efficient flexible job shop scheduling problem with processing- transportation composite robots
Framework of DSLCEA
Due to the EFJSP-PTCRs is a multi-objective optimization problem involving multiple sub-problems, the multi-objective CEA is a better option to solve this type of problems. The algorithm establishes multiple initial…
Continue Reading
-
Infrastructure deficits and informal settlements in sub-Saharan Africa
Keith, M. et al. A new urban narrative for sustainable development. Nat. Sustain. 6, 115–117 (2022).
Bettencourt, L. M. A. Introduction to Urban Science: Evidence and Theory of Cities as…
Continue Reading