Sparse Inversion with Iteratively Re-Weighted Least-Squares#

Least-squares inversion produces smooth models which may not be an accurate representation of the true model. Here we demonstrate the basics of inverting for sparse and/or blocky models. Here, we used the iteratively reweighted least-squares approach. For this tutorial, we focus on the following:

  • Defining the forward problem

  • Defining the inverse problem (data misfit, regularization, optimization)

  • Defining the paramters for the IRLS algorithm

  • Specifying directives for the inversion

  • Recovering a set of model parameters which explains the observations

import numpy as np
import matplotlib.pyplot as plt

from discretize import TensorMesh

from simpeg import (
    simulation,
    maps,
    data_misfit,
    directives,
    optimization,
    regularization,
    inverse_problem,
    inversion,
)

# sphinx_gallery_thumbnail_number = 3

Defining the Model and Mapping#

Here we generate a synthetic model and a mappig which goes from the model space to the row space of our linear operator.

nParam = 100  # Number of model paramters

# A 1D mesh is used to define the row-space of the linear operator.
mesh = TensorMesh([nParam])

# Creating the true model
true_model = np.zeros(mesh.nC)
true_model[mesh.cell_centers_x > 0.3] = 1.0
true_model[mesh.cell_centers_x > 0.45] = -0.5
true_model[mesh.cell_centers_x > 0.6] = 0

# Mapping from the model space to the row space of the linear operator
model_map = maps.IdentityMap(mesh)

# Plotting the true model
fig = plt.figure(figsize=(8, 5))
ax = fig.add_subplot(111)
ax.plot(mesh.cell_centers_x, true_model, "b-")
ax.set_ylim([-2, 2])
plot inv 2 inversion irls
(-2.0, 2.0)

Defining the Linear Operator#

Here we define the linear operator with dimensions (nData, nParam). In practive, you may have a problem-specific linear operator which you would like to construct or load here.

# Number of data observations (rows)
nData = 20

# Create the linear operator for the tutorial. The columns of the linear operator
# represents a set of decaying and oscillating functions.
jk = np.linspace(1.0, 60.0, nData)
p = -0.25
q = 0.25


def g(k):
    return np.exp(p * jk[k] * mesh.cell_centers_x) * np.cos(
        np.pi * q * jk[k] * mesh.cell_centers_x
    )


G = np.empty((nData, nParam))

for i in range(nData):
    G[i, :] = g(i)

# Plot the columns of G
fig = plt.figure(figsize=(8, 5))
ax = fig.add_subplot(111)
for i in range(G.shape[0]):
    ax.plot(G[i, :])

ax.set_title("Columns of matrix G")
Columns of matrix G
Text(0.5, 1.0, 'Columns of matrix G')

Defining the Simulation#

The simulation defines the relationship between the model parameters and predicted data.

sim = simulation.LinearSimulation(mesh, G=G, model_map=model_map)

Predict Synthetic Data#

Here, we use the true model to create synthetic data which we will subsequently invert.

# Standard deviation of Gaussian noise being added
std = 0.02
np.random.seed(1)

# Create a SimPEG data object
data_obj = sim.make_synthetic_data(true_model, noise_floor=std, add_noise=True)

Define the Inverse Problem#

The inverse problem is defined by 3 things:

  1. Data Misfit: a measure of how well our recovered model explains the field data

  2. Regularization: constraints placed on the recovered model and a priori information

  3. Optimization: the numerical approach used to solve the inverse problem

# Define the data misfit. Here the data misfit is the L2 norm of the weighted
# residual between the observed data and the data predicted for a given model.
# Within the data misfit, the residual between predicted and observed data are
# normalized by the data's standard deviation.
dmis = data_misfit.L2DataMisfit(simulation=sim, data=data_obj)

# Define the regularization (model objective function). Here, 'p' defines the
# the norm of the smallness term and 'q' defines the norm of the smoothness
# term.
reg = regularization.Sparse(mesh, mapping=model_map)
reg.reference_model = np.zeros(nParam)
p = 0.0
q = 0.0
reg.norms = [p, q]

# Define how the optimization problem is solved.
opt = optimization.ProjectedGNCG(
    maxIter=100, lower=-2.0, upper=2.0, maxIterLS=20, maxIterCG=30, tolCG=1e-4
)

# Here we define the inverse problem that is to be solved
inv_prob = inverse_problem.BaseInvProblem(dmis, reg, opt)

Define Inversion Directives#

Here we define any directiveas that are carried out during the inversion. This includes the cooling schedule for the trade-off parameter (beta), stopping criteria for the inversion and saving inversion results at each iteration.

# Add sensitivity weights but don't update at each beta
sensitivity_weights = directives.UpdateSensitivityWeights(every_iteration=False)

# Reach target misfit for L2 solution, then use IRLS until model stops changing.
IRLS = directives.UpdateIRLS(max_irls_iterations=40, f_min_change=1e-4)

# Defining a starting value for the trade-off parameter (beta) between the data
# misfit and the regularization.
starting_beta = directives.BetaEstimate_ByEig(beta0_ratio=1e0)

# Update the preconditionner
update_Jacobi = directives.UpdatePreconditioner()

# Save output at each iteration
saveDict = directives.SaveOutputEveryIteration(save_txt=False)

# Define the directives as a list
directives_list = [
    sensitivity_weights,
    IRLS,
    starting_beta,
    update_Jacobi,
    saveDict,
]

Setting a Starting Model and Running the Inversion#

To define the inversion object, we need to define the inversion problem and the set of directives. We can then run the inversion.

# Here we combine the inverse problem and the set of directives
inv = inversion.BaseInversion(inv_prob, directives_list)

# Starting model
starting_model = 1e-4 * np.ones(nParam)

# Run inversion
recovered_model = inv.run(starting_model)
Running inversion with SimPEG v0.23.1.dev10+gf697d2455

                    simpeg.InvProblem is setting bfgsH0 to the inverse of the eval2Deriv.
                    ***Done using the default solver Pardiso and no solver_opts.***

model has any nan: 0
=============================== Projected GNCG ===============================
  #     beta     phi_d     phi_m       f      |proj(x-g)-x|  LS    Comment
-----------------------------------------------------------------------------
x0 has any nan: 0
   0  1.77e+06  3.77e+03  1.02e-09  3.77e+03    1.99e+01      0
   1  8.83e+05  1.97e+03  3.62e-04  2.29e+03    1.93e+01      0
   2  4.42e+05  1.38e+03  8.52e-04  1.75e+03    1.90e+01      0   Skip BFGS
   3  2.21e+05  8.20e+02  1.75e-03  1.21e+03    1.78e+01      0   Skip BFGS
   4  1.10e+05  4.13e+02  3.05e-03  7.49e+02    1.66e+01      0   Skip BFGS
   5  5.52e+04  1.79e+02  4.51e-03  4.28e+02    1.43e+01      0   Skip BFGS
   6  2.76e+04  7.05e+01  5.86e-03  2.32e+02    1.30e+01      0   Skip BFGS
   7  1.38e+04  2.83e+01  6.90e-03  1.24e+02    1.08e+01      0   Skip BFGS
Reached starting chifact with l2-norm regularization: Start IRLS steps...
irls_threshold 1.2861779572001124
   8  1.38e+04  1.39e+01  1.04e-02  1.58e+02    1.50e+01      0   Skip BFGS
   9  1.38e+04  2.14e+01  1.10e-02  1.73e+02    1.39e+01      0
  10  8.14e+03  2.78e+01  1.16e-02  1.22e+02    9.56e+00      0   Skip BFGS
  11  5.24e+03  2.21e+01  1.29e-02  8.99e+01    8.47e+00      0
  12  3.38e+03  1.86e+01  1.37e-02  6.50e+01    8.73e+00      0   Skip BFGS
  13  4.46e+03  1.53e+01  1.41e-02  7.79e+01    1.38e+01      0   Skip BFGS
  14  5.89e+03  1.86e+01  1.30e-02  9.51e+01    1.52e+01      0
  15  3.70e+03  2.37e+01  1.17e-02  6.71e+01    1.01e+01      0
  16  2.32e+03  1.92e+01  1.12e-02  4.52e+01    1.01e+01      0
  17  3.16e+03  1.44e+01  1.05e-02  4.75e+01    1.12e+01      0
  18  4.18e+03  1.52e+01  8.76e-03  5.18e+01    1.24e+01      0
  19  5.27e+03  1.67e+01  7.29e-03  5.52e+01    1.24e+01      0
  20  6.39e+03  1.80e+01  6.09e-03  5.69e+01    1.22e+01      0
  21  7.75e+03  1.89e+01  5.08e-03  5.82e+01    1.23e+01      0
  22  9.40e+03  1.97e+01  4.23e-03  5.94e+01    1.26e+01      0
  23  1.14e+04  2.06e+01  3.51e-03  6.06e+01    1.30e+01      0
  24  1.38e+04  2.18e+01  2.95e-03  6.26e+01    1.38e+01      0
  25  8.72e+03  2.34e+01  2.45e-03  4.48e+01    1.17e+01      0
  26  5.50e+03  2.01e+01  2.53e-03  3.40e+01    1.34e+01      0
  27  3.47e+03  1.84e+01  2.28e-03  2.63e+01    1.05e+01      0
  28  4.54e+03  1.55e+01  2.13e-03  2.52e+01    1.07e+01      0
  29  5.96e+03  1.55e+01  1.83e-03  2.64e+01    1.75e+01      2
  30  7.87e+03  1.53e+01  1.41e-03  2.64e+01    1.34e+01      0
  31  1.05e+04  1.50e+01  1.06e-03  2.62e+01    1.47e+01      0
  32  1.40e+04  1.50e+01  8.11e-04  2.63e+01    1.42e+01      0
  33  1.88e+04  1.48e+01  6.33e-04  2.67e+01    1.42e+01      0
  34  2.52e+04  1.48e+01  5.09e-04  2.76e+01    1.45e+01      0
  35  3.37e+04  1.49e+01  4.22e-04  2.92e+01    1.45e+01      0
  36  4.48e+04  1.51e+01  3.49e-04  3.08e+01    1.45e+01      0
  37  5.91e+04  1.53e+01  2.70e-04  3.12e+01    1.44e+01      0
  38  7.78e+04  1.54e+01  2.20e-04  3.25e+01    1.45e+01      0
  39  1.02e+05  1.55e+01  1.82e-04  3.41e+01    1.46e+01      0   Skip BFGS
  40  1.34e+05  1.56e+01  1.52e-04  3.58e+01    1.46e+01      0
  41  1.74e+05  1.56e+01  1.26e-04  3.76e+01    1.45e+01      0
  42  2.27e+05  1.57e+01  1.05e-04  3.95e+01    1.46e+01      0
  43  2.95e+05  1.58e+01  8.72e-05  4.15e+01    1.48e+01      0
  44  3.82e+05  1.59e+01  7.26e-05  4.36e+01    1.51e+01      0
  45  4.94e+05  1.60e+01  6.05e-05  4.59e+01    1.53e+01      0
  46  6.36e+05  1.61e+01  5.05e-05  4.82e+01    1.56e+01      0
  47  8.17e+05  1.62e+01  4.22e-05  5.06e+01    1.58e+01      0
Reach maximum number of IRLS cycles: 40
------------------------- STOP! -------------------------
1 : |fc-fOld| = 0.0000e+00 <= tolF*(1+|f0|) = 3.7680e+02
1 : |xc-x_last| = 3.0506e-02 <= tolX*(1+|x0|) = 1.0010e-01
0 : |proj(x-g)-x|    = 1.5757e+01 <= tolG          = 1.0000e-01
0 : |proj(x-g)-x|    = 1.5757e+01 <= 1e3*eps       = 1.0000e-02
0 : maxIter   =     100    <= iter          =     48
------------------------- DONE! -------------------------

Plotting Results#

fig, ax = plt.subplots(1, 2, figsize=(12 * 1.2, 4 * 1.2))

# True versus recovered model
ax[0].plot(mesh.cell_centers_x, true_model, "k-")
ax[0].plot(mesh.cell_centers_x, inv_prob.l2model, "b-")
ax[0].plot(mesh.cell_centers_x, recovered_model, "r-")
ax[0].legend(("True Model", "Recovered L2 Model", "Recovered Sparse Model"))
ax[0].set_ylim([-2, 2])

# Observed versus predicted data
ax[1].plot(data_obj.dobs, "k-")
ax[1].plot(inv_prob.dpred, "ko")
ax[1].legend(("Observed Data", "Predicted Data"))

# Plot convergence
fig = plt.figure(figsize=(9, 5))
ax = fig.add_axes([0.2, 0.1, 0.7, 0.85])
ax.plot(saveDict.phi_d, "k", lw=2)

twin = ax.twinx()
twin.plot(saveDict.phi_m, "k--", lw=2)
ax.plot(
    np.r_[IRLS.metrics.start_irls_iter, IRLS.metrics.start_irls_iter],
    np.r_[0, np.max(saveDict.phi_d)],
    "k:",
)
ax.text(
    IRLS.metrics.start_irls_iter,
    0.0,
    "IRLS Start",
    va="bottom",
    ha="center",
    rotation="vertical",
    size=12,
    bbox={"facecolor": "white"},
)

ax.set_ylabel(r"$\phi_d$", size=16, rotation=0)
ax.set_xlabel("Iterations", size=14)
twin.set_ylabel(r"$\phi_m$", size=16, rotation=0)
  • plot inv 2 inversion irls
  • plot inv 2 inversion irls
Text(865.1527777777777, 0.5, '$\\phi_m$')

Total running time of the script: (0 minutes 41.155 seconds)

Estimated memory usage: 293 MB

Gallery generated by Sphinx-Gallery