Update README

package/samplers/moead/README.md

@@ -1,104 +1,74 @@
 ---
-author: Please fill in the author name here. (e.g., John Smith)
-title: Please fill in the title of the feature here. (e.g., Gaussian-Process Expected Improvement Sampler)
-description: Please fill in the description of the feature here. (e.g., This sampler searches for each trial based on expected improvement using Gaussian process.)
-tags: [Please fill in the list of tags here. (e.g., sampler, visualization, pruner)]
-optuna_versions: ['Please fill in the list of versions of Optuna in which you have confirmed the feature works, e.g., 3.6.1.']
+author: Hiroaki Natsume
+title: MOEA/D sampler
+description: Sampler using MOEA/D algorithm. MOEA/D stands for "Multi-Objective Evolutionary Algorithm based on Decomposition.
+tags: [sampler, multiobjective]
+optuna_versions: [4.0.0]
 license: MIT License
 ---
 
-<!--
-This is an example of the frontmatters.
-All columns must be string.
-You can omit quotes when value types are not ambiguous.
-For tags, a package placed in
-- package/samplers/ must include the tag "sampler"
-- package/visualilzation/ must include the tag "visualization"
-- package/pruners/ must include the tag "pruner"
-respectively.
-
----
-author: Optuna team
-title: My Sampler
-description: A description for My Sampler.
-tags: [sampler, 2nd tag for My Sampler, 3rd tag for My Sampler]
-optuna_versions: [3.6.1]
-license: "MIT License"
----
--->
-
-Please read the [tutorial guide](https://optuna.github.io/optunahub-registry/recipes/001_first.html) to register your feature in OptunaHub.
-You can find more detailed explanation of the following contents in the tutorial.
-Looking at [other packages' implementations](https://github.com/optuna/optunahub-registry/tree/main/package) will also help you.
-
 ## Abstract
 
-You can provide an abstract for your package here.
-This section will help attract potential users to your package.
+Sampler using MOEA/D algorithm. MOEA/D stands for "Multi-Objective Evolutionary Algorithm based on Decomposition.
 
-**Example**
-
-This package provides a sampler based on Gaussian process-based Bayesian optimization. The sampler is highly sample-efficient, so it is suitable for computationally expensive optimization problems with a limited evaluation budget, such as hyperparameter optimization of machine learning algorithms.
+This sampler is specialized for multiobjective optimization. The objective function is internally decomposed into multiple single-objective subproblems to perform optimization.
 
 ## Class or Function Names
 
-Please fill in the class/function names which you implement here.
-
-**Example**
-
-- GPSampler
+- MOEADSampler
 
 ## Installation
 
-If you have additional dependencies, please fill in the installation guide here.
-If no additional dependencies is required, **this section can be removed**.
-
-**Example**
-
-```shell
-$ pip install scipy torch
+```
+pip install scipy
 ```
 
 ## Example
 
-Please fill in the code snippet to use the implemented feature here.
-
-**Example**
-
 ```python
 import optuna
 import optunahub
 
+def objective(trial: optuna.Trial) -> tuple[float, float]:
+    x = trial.suggest_float("x", 0, 5)
+    y = trial.suggest_float("y", 0, 3)
 
-def objective(trial):
-  x = trial.suggest_float("x", -5, 5)
-  return x**2
+    v0 = 4 * x**2 + 4 * y**2
+    v1 = (x - 5) ** 2 + (y - 5) ** 2
+    return v0, v1
 
 
-sampler = optunahub.load_module(package="samplers/gp").GPSampler()
-study = optuna.create_study(sampler=sampler)
-study.optimize(objective, n_trials=100)
+if __name__ == "__main__":
+    population_size = 100
+    n_trials = 1000
+
+    mod = optunahub.load_module("samplers/moead")
+    sampler = mod.MOEADSampler(
+        population_size=population_size,
+        scalar_aggregation_func="tchebycheff",
+        n_neighbors=population_size // 10,
+    )
+    study = optuna.create_study(sampler=sampler)
+    study.optimize(objective, n_trials=n_trials)
 ```
 
 ## Others
 
-Please fill in any other information if you have here by adding child sections (###).
-If there is no additional information, **this section can be removed**.
+Comparison between Random, NSGAII and MOEA/D with ZDT1 as the objective function.
+See `compare_2objective.py` in moead directory for details.
+
+### Pareto Front Plot
 
-<!--
-For example, you can add sections to introduce a corresponding paper.
+| MOEA/D                      | NSGAII                       | Random                       |
+| --------------------------- | ---------------------------- | ---------------------------- |
+| ![MOEA/D](images/moead.png) | ![NSGAII](images/nsgaii.png) | ![Random](images/random.png) |
+
+### Compare
+
+![Compare](images/compare_pareto_front.png)
 
 ### Reference
-Takuya Akiba, Shotaro Sano, Toshihiko Yanase, Takeru Ohta, and Masanori Koyama. 2019.
-Optuna: A Next-generation Hyperparameter Optimization Framework. In KDD.
 
-### Bibtex
-```
-@inproceedings{optuna_2019,
-    title={Optuna: A Next-generation Hyperparameter Optimization Framework},
-    author={Akiba, Takuya and Sano, Shotaro and Yanase, Toshihiko and Ohta, Takeru and Koyama, Masanori},
-    booktitle={Proceedings of the 25th {ACM} {SIGKDD} International Conference on Knowledge Discovery and Data Mining},
-    year={2019}
-}
-```
--->
+Q. Zhang and H. Li,
+"MOEA/D: A Multiobjective Evolutionary Algorithm Based on Decomposition," in IEEE Transactions on Evolutionary Computation, vol. 11, no. 6, pp. 712-731, Dec. 2007,
+doi: 10.1109/TEVC.2007.892759.

package/samplers/moead/compare_2objectives.py

@@ -0,0 +1,54 @@
+import numpy as np
+import optuna
+import optunahub
+
+
+def objective(trial: optuna.Trial) -> tuple[float, float]:
+    # ZDT1
+    n_variables = 30
+
+    x = np.array([trial.suggest_float(f"x{i}", 0, 1) for i in range(n_variables)])
+    g = 1 + 9 * np.sum(x[1:]) / (n_variables - 1)
+    f1 = x[0]
+    f2 = g * (1 - (f1 / g) ** 0.5)
+
+    return f1, f2
+
+
+if __name__ == "__main__":
+    mod = optunahub.load_module("samplers/moead")
+
+    seed = 42
+    population_size = 100
+    n_trials = 10000
+    crossover = optuna.samplers.nsgaii.BLXAlphaCrossover()
+    samplers = [
+        optuna.samplers.RandomSampler(seed=seed),
+        optuna.samplers.NSGAIISampler(
+            seed=seed,
+            population_size=population_size,
+            crossover=crossover,
+        ),
+        mod.MOEADSampler(
+            seed=seed,
+            population_size=population_size,
+            n_neighbors=population_size // 5,
+            scalar_aggregation_func="tchebycheff",
+            crossover=crossover,
+        ),
+    ]
+    studies = []
+    for sampler in samplers:
+        study = optuna.create_study(
+            sampler=sampler,
+            study_name=f"{sampler.__class__.__name__}",
+            directions=["minimize", "minimize"],
+        )
+        study.optimize(objective, n_trials=n_trials)
+        studies.append(study)
+
+        optuna.visualization.plot_pareto_front(study).show()
+
+    m = optunahub.load_module("visualization/plot_pareto_front_multi")
+    fig = m.plot_pareto_front(studies)
+    fig.show()

package/samplers/moead/example.py

@@ -1,56 +1,33 @@
-import numpy as np
 import optuna
 import optunahub
 
-from moead import MOEAdSampler
 
+def objective(trial: optuna.Trial) -> tuple[float, float]:
+    x = trial.suggest_float("x", 0, 5)
+    y = trial.suggest_float("y", 0, 3)
 
-def objective(trial: optuna.Trial) -> tuple[float, float, float]:
-    n_variables = 3
-
-    x = np.array([trial.suggest_float(f"x{i}", 0, 1) for i in range(n_variables)])
-    n = 10
-    g = 100 * (n - 2) + 100 * np.sum((x - 0.5) ** 2 - np.cos(20 * np.pi * (x - 0.5)))
-
-    f1 = (1 + g) * x[0] * x[1]
-    f2 = (1 + g) * x[0] * (1 - x[1])
-    f3 = (1 + g) * (1 - x[0])
-
-    return f1, f2, f3
+    v0 = 4 * x**2 + 4 * y**2
+    v1 = (x - 5) ** 2 + (y - 5) ** 2
+    return v0, v1
 
 
 if __name__ == "__main__":
-    # mod = optunahub.load_module("samplers/moea_d")
-    # sampler = mod.MOEAdSampler()
-    seed = 42
-    population_size = 50
+    mod = optunahub.load_module("samplers/moead")
+
+    population_size = 100
     n_trials = 1000
     crossover = optuna.samplers.nsgaii.BLXAlphaCrossover()
-    samplers = [
-        optuna.samplers.RandomSampler(seed=seed),
-        # optuna.samplers.NSGAIIISampler(
-        #     seed=seed, population_size=population_size, crossover=crossover
-        # ),
-        MOEAdSampler(
-            seed=seed,
-            population_size=population_size,
-            n_neighbors=population_size // 5,
-            scalar_aggregation_func="tchebycheff",
-            crossover=crossover,
-        ),
-    ]
-    studies = []
-    for sampler in samplers:
-        study = optuna.create_study(
-            sampler=sampler,
-            study_name=f"{sampler.__class__.__name__}",
-            directions=["minimize", "minimize", "minimize"],
-        )
-        study.optimize(objective, n_trials=n_trials)
-        studies.append(study)
-
-    # optuna.visualization.plot_pareto_front(studies[0]).show()
-
-    m = optunahub.load_module("visualization/plot_pareto_front_multi")
-    fig = m.plot_pareto_front(studies)
-    fig.show()
+    sampler = mod.MOEADSampler(
+        population_size=population_size,
+        scalar_aggregation_func="tchebycheff",
+        n_neighbors=population_size // 10,
+        crossover=crossover,
+    )
+    study = optuna.create_study(
+        sampler=sampler,
+        study_name=f"{sampler.__class__.__name__}",
+        directions=["minimize", "minimize"],
+    )
+    study.optimize(objective, n_trials=n_trials)
+
+    optuna.visualization.plot_pareto_front(study).show()

package/samplers/moead/moead.py

@@ -51,12 +51,37 @@ def __init__(
         Args:
             seed:
                 Seed for random number generator.
-            population_size:
-                The number of individuals in the population.
-            T:
+
+            n_neighbors:
                 The number of the weight vectors in the neighborhood of each weight vector.
+                The larger this value, the more weight is applied to the exploration.
+
             scalar_aggregation_function:
-                The scalar aggregation function to use. The default is "weighted_sum". Other options are "tchebycheff" and "PBI".
+                The scalar aggregation function to use. The default is "tchebycheff". Other options is "weight_sum".
+
+            population_size:
+                Number of individuals (trials) in a generation.
+                ``population_size`` must be greater than or equal to ``crossover.n_parents``.
+                For :class:`~optuna.samplers.nsgaii.UNDXCrossover` and
+                :class:`~optuna.samplers.nsgaii.SPXCrossover`, ``n_parents=3``, and for the other
+                algorithms, ``n_parents=2``.
+
+            mutation_prob:
+                Probability of mutating each parameter when creating a new individual.
+                If :obj:`None` is specified, the value ``1.0 / len(parent_trial.params)`` is used
+                where ``parent_trial`` is the parent trial of the target individual.
+
+            crossover:
+                Crossover to be applied when creating child individuals.
+                For more information on each of the crossover method, please refer to
+                optuna crossover documentation.
+
+            crossover_prob:
+                Probability that a crossover (parameters swapping between parents) will occur
+                when creating a new individual.
+
+            swapping_prob:
+                Probability of swapping each parameter of the parents during crossover.
         """
         if population_size < 2:
             raise ValueError("`population_size` must be greater than or equal to 2.")