3 papers
cs.LG2026
Representing MAX functions using two-hidden-layer ReLU networks
Zhimao Wang, Amitabh Basu
We study exact representations of using two-hidden-layer ReLU neural networks. This problem has been studied in recent years in an attempt…
cs.LG2024
Towards Lower Bounds on the Depth of ReLU Neural Networks
Christoph Hertrich, Amitabh Basu, Marco Di Summa +1
We contribute to a better understanding of the class of functions that can be represented by a neural network with ReLU activations and a given architecture. Using techniques from…
cs.LG2024
On the power of graph neural networks and the role of the activation function
Sammy Khalife, Amitabh Basu
In this article we present new results about the expressivity of Graph Neural Networks (GNNs). We prove that for any GNN with piecewise polynomial activations, whose architecture s…