# The universal approximation theorem for complex-valued neural networks

arXiv: https://arxiv.org/abs/2012.03351 · HTML: https://mlsanity.com/paper/2012.03351/

## Key facts

- Cited by 28 papers in the map (data through 7 Oct 2026).
- First posted on arXiv 6 Dec 2020.

## Built on by (most cited first, 28 of 28)

- [Spherical Fourier Neural Operators: Learning Stable Dynamics on the Sphere](https://mlsanity.com/paper/2306.03838/) (2023)
- [Deep Learning Methods for Partial Differential Equations and Related Parameter Identification Problems](https://mlsanity.com/paper/2212.03130/) (2022)
- [Universal Approximation Theorem for Vector- and Hypercomplex-Valued Neural Networks](https://mlsanity.com/paper/2401.02277/) (2024)
- [Kolmogorov-Arnold Network Autoencoders](https://mlsanity.com/paper/2410.02077/) (2024)
- [Towards Understanding Theoretical Advantages of Complex-Reaction Networks](https://mlsanity.com/paper/2108.06711/) (2021)
- [Quantitative approximation results for complex-valued neural networks](https://mlsanity.com/paper/2102.13092/) (2021)
- [Scaling of neural-network quantum states for time evolution](https://mlsanity.com/paper/2104.10696/) (2021)
- [Invariant polynomials and machine learning](https://mlsanity.com/paper/2104.12733/) (2021)
- [Understanding Vector-Valued Neural Networks and Their Relationship with Real and Hypercomplex-Valued Neural Networks](https://mlsanity.com/paper/2309.07716/) (2023)
- [Theoretical Exploration of Flexible Transmitter Model](https://mlsanity.com/paper/2111.06027/) (2021)
- [Universal approximation with complex-valued deep narrow neural networks](https://mlsanity.com/paper/2305.16910/) (2023)
- [CoNO: Complex Neural Operator for Continous Dynamical Physical Systems](https://mlsanity.com/paper/2406.02597/) (2024)
- [A Survey of Deep Learning for Complex Speech Spectrograms](https://mlsanity.com/paper/2505.08694/) (2025)
- [Qualitative neural network approximation over R and C: Elementary proofs for analytic and polynomial activation](https://mlsanity.com/paper/2203.13410/) (2022)
- [Do stable neural networks exist for classification problems? -- A new view on stability in AI](https://mlsanity.com/paper/2401.07874/) (2024)
- [Approximation Rates and VC-Dimension Bounds for (P)ReLU MLP Mixture of Experts](https://mlsanity.com/paper/2402.03460/) (2024)
- [The neural networks with tensor weights and emergent fermionic Wick rules in the large-width limit](https://mlsanity.com/paper/2507.05303/) (2025)
- [Over-The-Air Extreme Learning Machines with XL Reception via Nonlinear Cascaded Metasurfaces](https://mlsanity.com/paper/2601.17749/) (2026)
- [On the Approximation and Complexity of Deep Neural Networks to Invariant Functions](https://mlsanity.com/paper/2210.15279/) (2022)
- [Selected aspects of complex, hypercomplex and fuzzy neural networks](https://mlsanity.com/paper/2301.00007/) (2022)
- [On the Computational Complexities of Complex-valued Neural Networks](https://mlsanity.com/paper/2310.13075/) (2023)
- [Neural Controlled Differential Equations with Quantum Hidden Evolutions](https://mlsanity.com/paper/2404.19673/) (2024)
- [On best approximation by multivariate ridge functions with applications to generalized translation networks](https://mlsanity.com/paper/2412.08453/) (2024)
- [Novel Complex-Valued Hopfield Neural Networks with Phase and Magnitude Quantization](https://mlsanity.com/paper/2507.00461/) (2025)
- [Algorithms and data structures for automatic precision estimation of neural networks](https://mlsanity.com/paper/2509.24607/) (2025)
- [Toward Complex-Valued Neural Networks for Waveform Generation](https://mlsanity.com/paper/2603.11589/) (2026)
- [Fourier Neural Operators with rank-1 lattice points and hyperbolic cross](https://mlsanity.com/paper/2606.08871/) (2026)
- [On the Expressive Power of Weight Quantization in Large Language Models](https://mlsanity.com/paper/2606.22249/) (2026)

## Abstract (arXiv)

We generalize the classical universal approximation theorem for neural networks to the case of complex-valued neural networks. Precisely, we consider feedforward networks with a complex activation function σ : mathbbC → mathbbC in which each neuron performs the operation mathbbCN → mathbbC, z mapsto σ(b + wT z) with weights w ∈ mathbbCN and a bias b ∈ mathbbC, and with σ applied componentwise. We completely characterize those activation functions σ for which the associated complex networks have the universal approximation property, meaning that they can uniformly approximate any continuous function on any compact subset of mathbbCd arbitrarily well. Unlike the classical case of real networks, the set of "good activation functions" which give rise to networks with the universal approximation property differs significantly depending on whether one considers deep networks or shallow networks: For deep networks with at least two hidden layers, the universal approximation property holds as long as σ is neither a polynomial, a holomorphic function, or an antiholomorphic function. Shallow networks, on the other hand, are universal if and only if the real part or the imaginary part of σ is not a polyharmonic function.