1 code implementation • 20 Feb 2024 • Nithin Chalapathi, Yiheng Du, Aditi Krishnapriyan
Our approach imposes the constraint over smaller decomposed domains, each of which is solved by an "expert" through differentiable optimization.
1 code implementation • 8 Dec 2023 • Yiheng Du, Nithin Chalapathi, Aditi Krishnapriyan
In contrast to current machine learning approaches which enforce PDE constraints by minimizing the numerical quadrature of the residuals in the spatiotemporal domain, we leverage Parseval's identity and introduce a new training strategy through a \textit{spectral loss}.
1 code implementation • 13 Dec 2019 • Archit Rathore, Nithin Chalapathi, Sourabh Palande, Bei Wang
To understand how such performance is achieved, we probe a trained deep neural network by studying neuron activations, i. e., combinations of neuron firings, at various layers of the network in response to a particular input.