Multi-Objective Evolutionary Semi-Supervised Learning for Resource-Aware Smart City Classification

Francisco José Sedeño; Jamal Toutouh

cs.LGOct 20, 2026

Multi-Objective Evolutionary Semi-Supervised Learning for Resource-Aware Smart City Classification

Authors: Francisco José Sedeño, Jamal Toutouh

Organizations: ITIS Software, University of Malaga, Malaga, Spain

Abstract

Smart-city applications often combine scarce labeled data with abundant unlabeled observations. Semi-supervised learning (SSL) exploits both labeled and unlabeled data to improve classification under limited supervision. This article presents MOEA-SSL, a multi-objective evolutionary approach for lightweight semi-supervised tabular classification that jointly optimizes predictive quality and feature usage. Experimental evaluation on three Smart City classification datasets, considering 1%, 5%, and 10% labeled data, shows that MOEA-SSL achieves competitive predictive performance compared with established lightweight SSL methods while significantly reducing feature usage relative to EA-SSL in all evaluated conditions. The computed Pareto fronts also provide lower-feature-use alternatives with limited degradation in classification performance.

Figures & tables

Component	Crossover	Mutation
Feature masks	Uniform exchange	Bit flip
Boolean genes	Uniform exchange	Boolean flip
Integer genes	Uniform exchange	Bounded $\pm 1$ step
Continuous genes	Arithmetic crossover	Bounded perturbation

Table 1: Variation operators inherited from EA-SSL.

Dataset	Domain	Samples	Number of features, $d$
Building Occupancy	Smart buildings	10,808	7
Grid Stability	Smart energy	10,000	12
Water Potability	Smart water	3,276	9

Table 2: Datasets used in the experimental evaluation.

Dataset	$\rho$	ST	HCo	EA-SSL	MOEA-SSL	Best Sup.
Building	1%	0.950[0.055]	0.881[0.040]	0.977[0.000]	0.977[0.000]	0.977[0.005] (SVM)
	5%	0.977[0.000]	0.947[0.014]	0.977[0.000]	0.977[0.000]	0.978[0.000] (SVM)
	10%	0.977[0.000]	0.948[0.014]	0.977[0.000]	0.977[0.001]	0.978[0.001] (SVM)
Grid	1%	0.683[0.076]	0.660[0.166]	0.746[0.032]	0.755[0.021]	0.737[0.040] (SVM)
	5%	0.757[0.020]	0.533[0.174]	0.764[0.017]	0.770[0.010]	0.829[0.016] (HGB)
	10%	0.782[0.007]	0.728[0.026]	0.761[0.010]	0.777[0.009] †	0.862[0.013] (HGB)

Table 3: Test MacroF1 across 30 runs. Values are median[inter-quartile range]. Bold indicates the highest median among the SSL methods. Best Sup. is the highest-median supervised approach and is descriptive only.

\dagger

denotes a significant MOEA-SSL vs. EA-SSL difference (Holm correction).

Dataset	$\rho$	MOEA-SSL	EA-SSL	Reduction	$p_{\mathrm{Holm}}$
Building	1%	0.429 (3/7)	0.857 (6/7)	50.0%	0.005
	5%	0.429 (3/7)	0.714 (5/7)	40.0%	0.010
	10%	0.286 (2/7)	0.714 (5/7)	60.0%	0.007
Grid	1%	0.667 (8/12)	0.833 (10/12)	20.0%	0.011
	5%	0.667 (8/12)	0.833 (10/12)	20.0%	0.010
	10%	0.750 (9/12)	0.917 (11/12)	18.2%	0.014

Table 4: Feature usage of the performance-oriented MOEA-SSL solution and EA-SSL. Values in parentheses are median selected features over

d

. Reduction is computed from the median FeatureRatio values. All paired FeatureRatio differences remain significant after Holm correction.

Figure 1: Test-space projection of validation-non-dominated solutions for representative median-HV runs. MOEA-SSL sets are selected using validation objectives only (vs. same sed EA-SSL).

Dataset	$\rho$	$z_{\mathrm{perf}}$ feat.	$z_{1\mathrm{pp}}$ feat.	Reduced	$F1_{\mathrm{perf}}$	$F1_{1\mathrm{pp}}$
Building	1%	3/7	2/7	16/30	0.9772	0.9772
	5%	3/7	2/7	16/30	0.9768	0.9772
	10%	2/7	2/7	18/30	0.9772	0.9772
Grid	1%	8/12	7/12	22/30	0.7554	0.7479
	5%	8/12	8/12	12/30	0.7701	0.7660
	10%	9/12	8/12	20/30	0.7770	0.7788

Table 5: Performance-retention analysis.

\mathbf{z}_{1\mathrm{pp}}

minimizes FeatureRatio within 0.01 validation RobustMacroF1 of

\mathbf{z}_{\mathrm{perf}}

. Feature counts and test MacroF1 are medians over 30 runs; “Reduced” counts runs in which

\mathbf{z}_{1\mathrm{pp}}

uses fewer features.

Dataset	$\rho$	HV	MOEA-SSL	EA-SSL	Ratio
Building	1%	0.701[0.100]	2487[1300]	1618[237]	1.54
	5%	0.705[0.138]	2847[1783]	1492[331]	1.91
	10%	0.706[0.139]	3329[1950]	1610[302]	2.07
Grid	1%	0.568[0.071]	1177[344]	1183[283]	1.00
	5%	0.573[0.040]	2434[686]	1392[269]	1.75
	10%	0.565[0.054]	1969[1512]	1427[250]	1.38

Table 6: Validation hypervolume and computational cost. Values are median[IQR] over the independent runs. Runtime is reported in seconds and Ratio denotes the median MOEA-SSL runtime divided by the median EA-SSL runtime.

Semi-Supervised Learning

Semi-supervised learning uses both labeled and unlabeled data to improve prediction when reliable annotations are scarce.

See all 164 papers

Cooperative Coevolution versus Monolithic Evolutionary Search for Semi-Supervised Tabular Classification
Apr 1, 2026Jamal ToutouhSemi-Supervised Learning Tabular ML
ITIS, Universidad de Málaga Spain
USE: Uncertainty Structure Estimation for Robust Semi-Supervised Learning
Feb 28, 2026Tsao-Lun Chen, Chien-Liang Liu, Tzu-Ming Harry Hsu +4Semi-Supervised Learning
National Taiwan University of Science and Technology, Taipei, Taiwan · National Taiwan University, Taipei, Taiwan · Chang Gung University, Taoyuan, Taiwan +3
Classifier-Dependent Benefits of Pseudo-Labeling for Semi-Supervised Android Malware Attribution
Aug 26, 2026Md Rafid Islam, Zahid Hasan, Hafiz Abdur RahmanPseudo-Labeling Malware Classification
Department of Electrical and Computer Engineering North South University Dhaka, Bangladesh

Apr 1, 2026J/K move · Enter open · S save

Cooperative Coevolution versus Monolithic Evolutionary Search for Semi-Supervised Tabular Classification

Jamal Toutouh

ITIS, Universidad de Málaga Spain

Semi-Supervised Learning Tabular ML Evolutionary Optimization Pseudo-Labeling

Open paper AskPDF arXiv

This paper studies semi-supervised tabular classification in the extreme low-label regime using lightweight base learners. The paper proposes a cooperative coevolutionary method (CC-SSL) that evolves (i) two feature-subset views and (ii) a pseudo-labeling policy, and compares it to a matched monolithic evolutionary baseline (EA-SSL) and three lightweight SSL baselines. Experiments on 25 OpenML datasets with labeled fractions {1%,5%,10%} evaluate test MacroF1 and accuracy, together with evolutionary and pseudo-label diagnostics. CC-SSL and EA-SSL achieve higher median test MacroF1 than the lightweight baselines, with the largest separations at 1% labeled data. Most CC-SSL vs. EA-SSL comparisons are statistical draws on final test performance. EA-SSL shows higher best-so-far fitness and higher diversity during search, while time-to-target is comparable and generations-to-target favors EA-SSL in several multiclass settings. Pseudo-label volume, ProbeDrop, and validation optimism show no significant differences between CC-SSL and EA-SSL under the shared protocol.

Abstract

Semi-Supervised Learning

Explore similar work

Cooperative Coevolution versus Monolithic Evolutionary Search for Semi-Supervised Tabular Classification

USE: Uncertainty Structure Estimation for Robust Semi-Supervised Learning

Classifier-Dependent Benefits of Pseudo-Labeling for Semi-Supervised Android Malware Attribution