EvoMO is a GPU-accelerated library for evolutionary multiobjective optimization (EMO) that leverages advanced tensorization techniques. By transforming key data structures and operations into tensor representations, EvoMO enables more efficient mathematical modeling and delivers significant performance improvements. Designed with scalability in mind, EvoMO can efficiently handle large populations and complex optimization tasks. Additionally, EvoMO includes MoRobtrol, a multiobjective robot control benchmark suite, providing a platform for testing tensorized EMO algorithms in real-world, black-box environments. EvoMO is a sister project of EvoX.
Note
To use the JAX version of EvoMO, you can switch to the v0.0.1-dev branch. This branch is fully compatible with EvoX version 0.9.0.
- EvoMO adopts a unified tensorization approach, restructuring EMO algorithms into tensor representations, enabling efficient GPU acceleration.
- Supports tensorized implementations of NSGA-II, NSGA-III, MOEA/D, RVEA, HypE, and more, achieving up to 1113× speedup while preserving solution quality.
- Handles large populations, scaling to hundreds of thousands for complex optimization tasks, ensuring scalability for real-world applications.
- Includes MoRobtrol, a multiobjective robot control benchmark, for testing tensorized EMO algorithms in challenging black-box environments.
- After installation, you can directly import EvoMO algorithms using
import evox, making it seamless to access both EvoX and EvoMO algorithms with a unified interface.
To install EvoMO, you need to install EvoX first.
- Install EvoX:
pip install evox- Install EvoMO:
pip install evomoFor the latest development version, you can install from the source:
git clone https://github.com/EMI-Group/evomo.git
cd evomo
pip install -e.Solve the DTLZ2 problem using the TensorMOEA/D algorithm:
import time
import torch
from evox.algorithms import TensorMOEAD
from evox.metrics import igd
from evox.problems.numerical import DTLZ2
from evox.workflows import StdWorkflow
if __name__ == "__main__":
torch.set_default_device("cuda" if torch.cuda.is_available() else "cpu")
algo = TensorMOEAD(pop_size=100, n_objs=3, lb=-torch.zeros(12), ub=torch.ones(12))
prob = DTLZ2(m=3)
pf = prob.pf()
workflow = StdWorkflow(algo, prob)
workflow.init_step()
jit_state_step = torch.compile(workflow.step)
t = time.time()
for i in range(100):
print(i)
jit_state_step()
fit = workflow.algorithm.fit
fit = fit[~torch.any(torch.isnan(fit), dim=1)]
print(f"Generation {i + 1} IGD: {igd(fit, pf)}")
print(f"Total time: {time.time() - t} seconds")Solve the MoSwimmer problem in MoRobtrol using the TensorMOEA/D algorithm:
import time
import torch
import torch.nn as nn
from evox.algorithms import TensorMOEAD
from evox.problems.neuroevolution import MoRobtrol
from evox.utils import ParamsAndVector
from evox.workflows import EvalMonitor, StdWorkflow
class SimpleMLP(nn.Module):
def __init__(self):
super(SimpleMLP, self).__init__()
self.features = nn.Sequential(nn.Linear(8, 4), nn.Tanh(), nn.Linear(4, 2))
def forward(self, x):
return torch.tanh(self.features(x))
def setup_workflow(model, pop_size, max_episode_length, num_episodes, device):
adapter = ParamsAndVector(dummy_model=model)
model_params = dict(model.named_parameters())
pop_center = adapter.to_vector(model_params)
lower_bound = torch.full_like(pop_center, -5)
upper_bound = torch.full_like(pop_center, 5)
problem = MoRobtrol(
policy=model,
env_name="mo_swimmer",
max_episode_length=max_episode_length,
num_episodes=num_episodes,
pop_size=pop_size,
device=device,
num_obj=2,
observation_shape=8,
obs_norm=torch.tensor([5.0, 1e-6, 1e6], device=device),
)
algorithm = TensorMOEAD(
pop_size=pop_size, lb=lower_bound, ub=upper_bound, n_objs=2, device=device
)
monitor = EvalMonitor(device=device)
workflow = StdWorkflow(
algorithm=algorithm,
problem=problem,
monitor=monitor,
opt_direction="max",
solution_transform=adapter,
device=device,
)
return workflow
def run_workflow(workflow, compiled=True, generations=10):
workflow.init_step()
step_function = torch.compile(workflow.step) if compiled else workflow.step
for index in range(generations):
print(f"In generation {index}:")
t = time.time()
step_function()
print(f"\tFitness: {-workflow.algorithm.fit}.")
print(f"\tTime elapsed: {time.time() - t: .4f}(s).")
if __name__ == "__main__":
device = "cuda" if torch.cuda.is_available() else "cpu"
model = SimpleMLP().to(device)
workflow = setup_workflow(model, 12, 100, 2, device)
run_workflow(workflow)- Hao Li, Zhenyu Liang, and Ran Cheng, “GPU-accelerated evolutionary many-objective optimization using tensorized NSGA-III,” in IEEE Congress on Evolutionary Computation, 2025. [📄 Paper] | [🧐 Read More]
- Zhenyu Liang, Tao Jiang, Kebin Sun, and Ran Cheng, “GPU-accelerated evolutionary multiobjective optimization using tensorized RVEA,” in Proceedings of the Genetic and Evolutionary Computation Conference, 2024, pp. 566–575. [📄 Paper] | [🧐 Read More]
We welcome contributions and look forward to your feedback!
- Engage in discussions and share your experiences on GitHub Issues.
- Join our QQ group (ID: 297969717).
If you use EvoMO in your research and want to cite it in your work, please use:
@article{evomo,
title = {Bridging Evolutionary Multiobjective Optimization and {GPU} Acceleration via Tensorization},
author = {Liang, Zhenyu and Li, Hao and Yu, Naiwei and Sun, Kebin and Cheng, Ran},
journal = {IEEE Transactions on Evolutionary Computation},
year = 2025,
doi = {10.1109/TEVC.2025.3555605}
}