Challenges for non-selfadjoint spectral problems in analysis and computation

Challenges for non-selfadjoint spectral problems in analysis and computation

🎙 Christiane Tretter 👥 8K 📅 August 18, 2026 ⏱ 45 min 👁 7 📄 original study 🧭 2026-08-18
Available in: English (current) Français

Keywords

essential numerical rangenon-selfadjointspectral pollutionspectral approximationoperator theory

Summary

The talk addresses challenges in non-selfadjoint spectral problems, focusing on the instability of spectra and the phenomena of spectral pollution and invisibility in numerical approximations. The speaker introduces the essential numerical range as a tool to capture spectral pollution and provides rigorous results for projection methods and domain truncation. The talk includes motivating examples from advection-diffusion operators and Maxwell’s equations with non-vanishing conductivity. Key results include the characterization of spurious eigenvalues and the enclosure of the essential spectrum. The speaker also discusses stability properties and resolvent estimates. The talk concludes with applications to domain truncation for Maxwell’s equations, providing spectral enclosures and resolvent bounds.

104 words

Critical Evaluation

Value of the Information & Strength of the Argument

The talk presents original research with rigorous mathematical proofs. The argumentation is solid, building from motivating examples to abstract results and applications. The speaker clearly explains the limitations of existing methods and demonstrates how the essential numerical range addresses them. The results are sharp, with examples showing optimality. The talk bridges theory and computation, offering practical guidance for numerical methods.

Scientific Rigor, Source Quality, Title Accuracy

The talk is scientifically rigorous, with precise definitions and theorems. The speaker cites relevant literature, including works by Fillmore, Stampfli, and Williams, and her own joint papers. The title accurately reflects the content. The talk is part of a workshop at the Isaac Newton Institute, ensuring high academic standards.

124 words

Title / Content Match

The title accurately reflects the content, which focuses on challenges and solutions for non-selfadjoint spectral problems, combining analysis and computation.

Quality & Reliability

9/10

Talk by a leading expert in operator theory, presenting rigorous mathematical results with proofs and references to published work. The content is highly technical and precise, with clear definitions and theorems. The speaker is a professor at a renowned university, and the talk is part of a prestigious workshop at the Isaac Newton Institute.

Key Moments

Cited Sources

Concurring Sources

  • Fillmore, Stampfli, Williams - Essential numerical range — Original definition for bounded operators.
  • Tretter, Malamud, Berkolaiko - JFA paper — Joint work on essential numerical range for unbounded operators.

Contribution & Novelties

The talk presents novel results on the essential numerical range for unbounded operators, including equivalent definitions, stability properties, and applications to spectral approximation. It provides a rigorous framework to understand and mitigate spectral pollution in numerical methods.

Pour aller plus loin :

  • Essential spectrum — Background on essential spectrum variants.
  • Numerical range — Definition and properties.
  • Spectral pollution — Overview of the phenomenon.
  • Non-self-adjoint operator — General context.

68 words

Radar Profile

The radar profile shows very high scores in information quality and technical level, with slightly lower but still high scores in quantity and reliability. This indicates a dense, expert-level talk with strong scientific content, though the quantity of information is limited by the talk's duration.

Reliability 9/10