Results 11 to 20 of about 6,811,260 (293)
Expressive power of parametrized quantum circuits
Physical Review Research, 2020 Parametrized quantum circuits (PQCs) have been broadly used as a hybrid quantum-classical machine learning scheme to accomplish generative tasks. However, whether PQCs have better expressive power than classical generative neural networks, such as ...
Yuxuan Du +3 more
doaj +2 more sources
On the Expressive Power of String Constraints
Proceedings of the ACM on Programming Languages, 2023 We investigate properties of strings which are expressible by canonical types of string constraints. Specifically, we consider a landscape of 20 logical theories, whose syntax is built around combinations of four common elements of string constraints: language membership (e.g.
Joel D. Day +3 more
openaire +3 more sources
Rethinking the Expressive Power of GNNs via Graph Biconnectivity [PDF]
International Conference on Learning Representations, 2023 Designing expressive Graph Neural Networks (GNNs) is a central topic in learning graph-structured data. While numerous approaches have been proposed to improve GNNs in terms of the Weisfeiler-Lehman (WL) test, generally there is still a lack of deep ...
Bohang Zhang +3 more
semanticscholar +1 more source
On the Expressive Power of Geometric Graph Neural Networks [PDF]
International Conference on Machine Learning, 2023 The expressive power of Graph Neural Networks (GNNs) has been studied extensively through the Weisfeiler-Leman (WL) graph isomorphism test. However, standard GNNs and the WL framework are inapplicable for geometric graphs embedded in Euclidean space ...
Chaitanya K. Joshi, Simon V. Mathis
semanticscholar +1 more source
The Expressive Power of Graph Neural Networks: A Survey [PDF]
IEEE Transactions on Knowledge and Data Engineering, 2023 Graph neural networks (GNNs) are effective machine learning models for many graph-related applications. Despite their empirical success, many research efforts focus on the theoretical limitations of GNNs, i.e., the GNNs expressive power.
Bingxue Zhang +6 more
semanticscholar +1 more source
The expressive power of pooling in Graph Neural Networks [PDF]
Neural Information Processing Systems, 2023 In Graph Neural Networks (GNNs), hierarchical pooling operators generate local summaries of the data by coarsening the graph structure and the vertex features.
F. Bianchi, Veronica Lachi
semanticscholar +1 more source
Effect of data encoding on the expressive power of variational quantum-machine-learning models [PDF]
Physical Review A, 2020 Quantum computers can be used for supervised learning by treating parametrised quantum circuits as models that map data inputs to predictions. While a lot of work has been done to investigate practical implications of this approach, many important ...
M. Schuld +2 more
semanticscholar +1 more source
On Structural Expressive Power of Graph Transformers [PDF]
Knowledge Discovery and Data Mining, 2023 Graph Transformer has recently received wide attention in the research community with its outstanding performance, yet its structural expressive power has not been well analyzed.
Wenhao Zhu +4 more
semanticscholar +1 more source
Neural tensor contractions and the expressive power of deep neural quantum states [PDF]
Physical review B, 2021 We establish a direct connection between general tensor networks and deep feed-forward artificial neural networks. The core of our results is the construction of neural-network layers that efficiently perform tensor contractions, and that use commonly ...
Or Sharir, A. Shashua, Giuseppe Carleo
semanticscholar +1 more source
Learning General Optimal Policies with Graph Neural Networks: Expressive Power, Transparency, and Limits [PDF]
International Conference on Automated Planning and Scheduling, 2021 It has been recently shown that general policies for many classical planning domains can be expressed and learned in terms of a pool of features defined from the domain predicates using a description logic grammar.
Simon Ståhlberg +2 more
semanticscholar +1 more source