Abstract:We discuss the growth of the Lebesgue constants for polynomial interpolation at Fekete points for fixed degree (one) and varying dimension, and underlying set $K\subset \R^d$ a simplex, ball or cube.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to explore how the Lebesgue constant grows with the change of dimension in Fekete point interpolation of a fixed degree (degree 1), especially its performance on different geometries (such as simplex, sphere and cube).
Specifically, the paper studies the following problems:
1. **Growth trend of the Lebesgue constant**: For a given geometry (such as simplex, sphere and cube), how does the Lebesgue constant \(\Lambda_1(X; K)\) change when the dimension \(d\) increases. In particular, for polynomial interpolation of degree 1, that is, \(n = 1\), how does the Lebesgue constant of the Fekete point set \(X\) change with the change of dimension.
2. **Comparison of different geometries**: The paper compares the growth of the Lebesgue constant on simplex, sphere and cube, and points out that the simplex has unique properties, that is, its Lebesgue constant does not increase with the increase of dimension, while the Lebesgue constants on the sphere and cube will increase with the increase of dimension.
3. **Theoretical verification and boundary estimation**: Through theoretical derivation and mathematical proof, specific upper bounds of the Lebesgue constant under different geometries are given, and the correctness of these upper bounds is verified.
### Main contributions of the paper
- **Lebesgue constant on simplex**: It is proved that for any dimension \(d\), the Lebesgue constant \(\Lambda_1(X; K)\) on simplex is always equal to 1, that is:
\[
\Lambda_1(X; K)=1, \quad \forall d\geq1.
\]
- **Lebesgue constant on sphere**: The upper and lower bound estimates of the Lebesgue constant on sphere are given:
\[
\sqrt{d}\leq\Lambda_1\leq\sqrt{d + 1}.
\]
- **Lebesgue constant on cube**: Using the properties of Hadamard matrices, the maximum value of the Lebesgue constant on cube in specific dimensions is discussed, and it is proved that in some special dimensions, the Lebesgue constant can reach \(\sqrt{d + 1}\).
### Conclusion
By analyzing in detail the growth of the Lebesgue constant on different geometries, the paper reveals the unique properties of simplex and provides an important theoretical basis for polynomial interpolation in high - dimensional spaces.