Graph neural networks with configuration cross-attention for tensor compilers by Dmitrii Khizbullin, Eduardo Rocha de Andrade, Thanh Hau Nguyen, Matheus Pedroza Ferreira, David R. Pugh

“…A neural network inference workload can be represented as a computational graph with nodes as operators transforming multidimensional tensors. …”
Get full text

Iterative Linear Quadratic Optimization for Nonlinear Control: Differentiable Programming Algorithmic Templates by Roulet, Vincent, Srinivasa, Siddhartha, Fazel, Maryam, Harchaoui, Zaid

“…In a differentiable programming framework, this information, such as gradients, can be automatically derived from the computational graph. We explore how nonlinear control algorithms, often employing linear and/or quadratic approximations, can be effectively cast within this framework. …”
Get full text

DropKAN: Dropout Kolmogorov–Arnold Networks by Mohammed Ghaith Altarabichi

“…We propose DropKAN (Dropout Kolmogorov—Arnold Networks), a regularization method that introduces dropout masks at the edge level within Kolmogorov—Arnold Networks (KANs) layers, randomly masking a subset of activation outputs in the computation graph. Forward pass analysis reveals that DropKAN, when combined with scaling, accurately preserves the expected output signal magnitude in line with theoretical expectations. …”
Get full text

A multi-modal graph-based framework for Alzheimer’s disease detection by Najmeh Mashhadi, Razvan Marinescu

“…In our directed computational graph, datasets are represented as nodes $$n_i$$ , and deep learning (DL) models are represented as directed edges $$n_i \rightarrow n_j$$ , allowing us to model complex image-processing pipelines $$n_1 \rightarrow n_2 \rightarrow n_3... …”
Get full text

PROBABLE CALCULATION OF BUILDING SEISMIC RESISTANCE ON KINEMATIC SUPPORT by H. M. Muselemov, O. M. Ustarkhanov, A. K. Yusupov

“…Based on the results of numerical experiments conducted on a computer, graphs of the reliability of seismic stability of the building in earthquakes. …”
Get full text