---
title: AAA-Lawson Rational Approximation
url: https://www.emergentmind.com/topics/aaa-lawson
type: topic
---

# AAA-Lawson Rational Approximation

AAA-Lawson is a two-phase numerical algorithm for uniform-norm rational approximation of functions on real intervals or complex domains. It combines the adaptive Antoulas–Anderson (AAA) barycentric rational fitting, which quickly generates a near-minimax approximant, with a nonconvex Lawson-style iteratively reweighted least squares (IRLS) refinement phase that drives the approximation toward true Chebyshev optimality (minimizing the ℓ_∞ norm of the error) [1908.06001, 2308.06991, 2305.03677, 2511.04404]. The method is applicable to both real and complex data, arbitrary degrees, and diverse geometric domains, and is central to modern rational minimax approximation toolkits such as Chebfun.  

## 1. Problem Setting and Motivation

Given sample points $\{z_j\}_{j=1}^M \subset \mathbb{C}$, data values $f_j = f(z_j)$, and a target rational type (typically $(n,n)$ or $(m,n)$), the objective is
\[
\min_{r \in \mathcal R_{m,n}} \max_{1 \leq j \leq M} |f_j - r(z_j)|
\]
where $\mathcal R_{m,n}$ denotes the set of rational functions $r(z)=p(z)/q(z)$ of numerator degree $\leq m$ and denominator degree $\leq n$ with $q(z_j)\neq 0$.

Traditional Remez-type algorithms are not robust for complex data or arbitrary domains. AAA-Lawson overcomes these limitations by leveraging the numerically stable barycentric AAA representation and enhancing it with IRLS to approximate the minimax solution even for highly ill-conditioned or nonclassical geometries [1908.06001].

## 2. Algorithmic Framework: AAA Phase and Lawson Phase

### AAA Phase

The adaptive Antoulas–Anderson (AAA) algorithm constructs a rational function in barycentric form
\[
r(z) = \frac{\sum_{k=0}^m \alpha_k f(t_k)/(z-t_k)}{\sum_{k=0}^m \alpha_k/(z-t_k)}
\]
by greedily selecting “support points” $\{t_k\}$ at maximum residuals and solving a sequence of least-squares problems using the SVD of a Loewner/Cauchy-type matrix. This phase yields a rational approximant that is generally within 1–2 digits of the true minimax error, but typically not equioscillatory, and so not certifiably optimal in the Chebyshev sense [1908.06001, 2308.06991].

### Lawson Phase (IRLS)

With the support set $\{t_k\}$ and initial weights $\alpha_k$ fixed, the AAA-Lawson phase addresses
\[
\min_{\alpha,\beta}\; \max_{1 \leq j \leq M} \left| f_j - \frac{\sum_{k=0}^n \alpha_k/(z_j-t_k)}{\sum_{k=0}^n \beta_k/(z_j-t_k)} \right|
\]
via iteratively reweighted least squares: at each step, solve a weighted least-squares problem minimizing the sum of squared residuals penalized by inverse magnitudes of the current residuals, iterate weights as $w_j \leftarrow 1/ ( |f_j - r(z_j)|+\epsilon )$, and continue until the error stabilizes or a prescribed maximum iteration count is reached [2305.03677, 2511.04404].

Pseudocode excerpt ([2305.03677], [2308.06991]):
```matlab
for t = 1:maxLawsonSteps
   W = 1 ./ (abs(res) + eps);
   A = (F - f(S)) ./ (z - S);
   AW = bsxfun(@times, sqrt(W), A);
   [~,~,V] = svd(AW,0);
   w = V(:,end);
   r = barycentric_eval(S, w, f(S), z);
   res = F - r(z);
end
```

## 3. Mathematical Structure and Variants

The AAA-Lawson approach can be analyzed from multiple perspectives:

- **Weighted Least Squares**: Each IRLS step solves
  \[
  \min_{\|w\|_2=1} \left\| W^{1/2} A w \right\|_2
  \]
  where $A$ is the barycentric Loewner/Cauchy matrix and $W=\mathrm{diag}(1/|e_j|+\epsilon)$ penalizes large errors [1908.06001, 2305.03677].

- **Convex Duality**: The IRLS process is interpreted as a block-coordinate ascent on the dual of a linearized minimax problem. Specifically, for complex rational minimax, the dual variable $w$ lies in a simplex, and maximizing the dual functional under Ruttan’s condition certifies global optimality [2308.06991].

- **Lawson Parameter ($\beta$)**: Modifies weight update as $w_j^{(k+1)} \propto w_j^{(k)} |f_j - r^{(k)}(x_j)|^\beta$. Choosing $\beta=1$ yields classical Lawson; smaller $\beta$ adds damping and may improve stability [2308.06991].

- **Continuum AAA-Lawson**: In domains requiring dense adaptive refinement (e.g., near singularities on $\mathbb{R}$ or $\mathbb{T}$), AAA-Lawson operates on adaptively discretized grids and employs high-resolution sampling at each step to stabilize the error curve [2305.03677].

## 4. Numerical Performance and Comparative Analysis

AAA-Lawson’s performance has been thoroughly benchmarked against the Loewner framework, RKFIT, and traditional AAA:

| Method           | Achievable $\ell_\infty$ error | Computational Cost       | Artifacts          |
|------------------|-------------------------------|-------------------------|--------------------|
| AAA              | Fast, not near-minimax        | Mild, nearly linear     | Spurious poles     |
| AAA-Lawson       | Near-minimax                  | $O(r^3)$, costly at high $r$ | Spurious complex structure possible |
| Loewner Framework| Near-minimax or better        | Cost flat in $r$        | Structure-preserving   |

- AAA-Lawson achieves uniform errors down to $10^{-12}$–$10^{-13}$ in double precision for moderate degrees and grids of size $N\lesssim 2000$, outperforming AAA alone [1908.06001, 2511.04404].
- For high-degree rational approximants ($r\geq 20$), the Loewner framework can be both faster and more accurate, especially with respect to computational cost, and avoids introducing non-physical complex coefficients in real-data problems.
- For applications demanding equioscillation or strict Chebyshev optimality, AAA-Lawson provides a practical compromise between direct optimization and numerical tractability.

## 5. Convergence, Limitations, and Certificates of Optimality

- **Convergence**: Empirically, AAA-Lawson shows rapid, often linear convergence of the maximum residual, typically achieving one additional digit of accuracy every 5–10 steps until stagnation at machine precision [1908.06001, 2305.03677].
- **Certification**: While AAA-Lawson reliably pushes the error toward equioscillation, only “dual-Lawson” (convex dual) variants as in [2308.06991] can rigorously certify attainment of the true minimax error, as monitored by the duality gap.
- **Non-monotonicity**: If support set selection or weight damping ($\beta$) is suboptimal, the error or dual objective may momentarily worsen before improving.
- **Artifacts**: For high-degree or large-scale problems, spurious complex-conjugate pole/zero pairs may appear in problems with real data due to lack of symmetry constraints in the Lawson phase.
- **Computational cost**: The Lawson phase is $O(L r^2 N)$ for $L$ steps, where $r$ is the degree and $N$ the grid size; for large $r$ this becomes prohibitive.

## 6. Implementation and Applications

AAA-Lawson is implemented in open-source software suites such as Chebfun. Maximal practical benefit is achieved when:
- The initial AAA approximant is already within a factor $10$–$100$ of minimax, ensuring a well-conditioned support set.
- The number of Lawson steps is chosen modest (e.g., $10$–$20$).
- Applications include spectrally accurate function approximation, model reduction, physical system identification, and numerical solution of inverse Zolotarev problems [2511.04404, 2305.03677].

In summary, AAA-Lawson is a robust, widely utilized approach for rational minimax approximation, combining the adaptivity of AAA barycentric sampling with the uniform-norm refinement of Lawson IRLS. It is most suitable for medium-scale or moderate-degree problems where near-optimal uniform error and equioscillation are critical, while alternative methods such as Loewner or convex-dual approaches may be preferable in large-scale or certifiably minimax settings.

Source: https://www.emergentmind.com/topics/aaa-lawson