arXiv:2605.24031v1 Announce Type: cross Abstract: We study the reconstruction of implied volatility surfaces from sparse and noisy option quotes using deep learning models under no-arbitrage constraints. We compare multiple neural architectures, including multilayer perceptrons, convolutional networks, U-Nets, variational autoencoders, and Transformer-based models against classical SVI parameterizations on option market data. Results show that Transformer and U-Net architectures achieve strong reconstruction accuracy, particularly under sparse observation regimes, while soft arbitrage penaltie
The increasing availability of advanced deep learning architectures and computational power makes this research timely, enabling more sophisticated financial modeling than previously possible.
This development could significantly enhance the accuracy and efficiency of financial risk management and derivative pricing, particularly in volatile or illiquid markets.
The ability to reconstruct volatility surfaces more accurately under limited data conditions changes how financial institutions can price complex options and manage market exposure.
- · Quantitative hedge funds
- · Investment banks
- · AI/ML financial tech providers
- · Option market makers
- · Traditional statistical modeling firms
- · Manual volatility traders
Financial institutions will integrate these deep learning models to improve their derivative pricing and risk analysis.
Increased adoption of these models could lead to more robust and less arbitrage-prone option markets.
The democratization of advanced volatility surface reconstruction could lower barriers to entry for sophisticated quantitative trading, potentially increasing market efficiency and competition.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG