Library · Machine learning & AI
What Supervised & Unsupervised Learning got wrong, from 37 theses
Summary written by a model from the records below: Across the evaluated studies, unsupervised learning methods frequently struggled with class separation, semantic misalignment, and complex representation tuning. At the same time, supervised models were vulnerable to noisy labels, loss distortion, and underperforming simpler baselines or heuristic alternatives. These records come from graduate theses, 2021 to 2026, each labeled with its degree where the repository states it, and each links to its thesis. They were extracted by language models reading the full text, so treat each as a lead to read, not a verdict.
Plus 4 problems the authors left open, listed at the end.
Failure modes, by theses reporting them
Select a mode to jump to its records.
Unsupervised clustering and dimensionality reduction fail to separate overlapping signals or underlying classes
Dimensionality reduction and clustering algorithms frequently failed to create distinct clusters that separate abnormal samples, disease states, sensory traits, or bacterial serovars due to heavy spectral and temporal overlap. Researchers rejected these unsupervised methods because they obscured fine-grained chemical gradients, struggled with ambient acoustic noise, and could not untangle complex boundaries.
Tried and failed
unsupervised acoustic clustering applied to vessel noise identification. Reason: algorithm grouped co-occurring impulsive sounds instead due to shallow-water low-frequency propagation limits
Understanding the Physical Environment of a French Polynesian Coral Reef Ecosystem Using Ocean Modeling, Acoustics, And Machine Learning · Georgia Tech · PhD
Tried and failed
unsupervised consensus clustering applied to plasma proteomic profiles in disease cohorts. Outcome: no signal. Reason: failed to yield statistically significant or distinct patient clusters
Using multi-omics to investigate the host response to COVID-19 · Imperial · PhD
Tried and failed
unsupervised dimensionality reduction for clustering applied to biomarker lipidomics profiles across timepoints. Outcome: no signal. Reason: unsupervised methods failed to clearly separate disease and control classes across combined timepoints
Metabolomics and Machine Learning for Early-Stage Cancer Diagnosis · Georgia Tech · PhD
Tried and failed
unsupervised clustering of mass spectrometry profiles applied to meat sensory and aging traits. Outcome: no signal. Reason: clustering failed to separate samples across sensory responses, tenderness metrics, or product age
Evaluating the Ability of Rapid Evaporative Ionization Mass Spectrometry to Predict the Palatability of Long Aged Beef Cuts · Texas Tech · Master's
Tried and failed
unsupervised clustering on mass spectrometry spectra applied to bacterial serovar differentiation. Outcome: no signal. Reason: significant spectral overlap and poor separation among closely related classes
Considered and rejected
Considered and rejected: Rejected pure statistical stream clustering (WCDS, FISHDBC) and unsupervised Autoencoders alone due to inability to untangle overlapping temporal boundaries in IMU data.
2022: A Computational Odyssey - Towards a Deeper Understanding of Clustering Streaming Human Activity Recognition Data · Queen's University · Master's
Considered and rejected
Considered and rejected: Rejected using the unsupervised SPICE/clustering detector pipeline for sperm whale daily presence due to clustering difficulty separating clicks from low-frequency (<10 kHz) anthropogenic and vessel noise.
Passive Acoustic Monitoring of Whales and Ocean Ambient Sound to Inform Management · Cornell · PhD
Considered and rejected
Considered and rejected: Rejected relying solely on unsupervised sample clustering/PCA because it obscured fine-grained microbial sensitivity to specific chemical gradients
MICROBIAL GENES, GENOMES AND TAXA ASSOCIATED WITH KEY ASPECTS OF PATHOGENESIS AND BIOGEOCHEMICAL CYCLES · Johns Hopkins · PhD
Considered and rejected
Considered and rejected: Rejected simple K-means clustering (with purely unsupervised centroids) for algospeak topic categorization due to inability to generate clear, stable, and balanced clusters.
Checking, Moderating, Adapting: Collective Engagement for Digital Integrity · Cornell · PhD
Unsupervised topics and clusters fail to align with predefined expert taxonomies or interpretable concepts
Unsupervised topic models, clustering, and representation learning often discovered groupings that failed to match target expert taxonomies, meaningful symbolic concepts, or social categories. Practitioners rejected fully autonomous clustering because it required manual labeling, lacked user control over thematic detail, and suffered from opaque hyperparameter tuning.
Tried and failed
unsupervised topic modeling and clustering applied to document classification into expert taxonomy. Outcome: worse than baseline. Reason: discovered clusters failed to align with predefined expert taxonomies without reference information
Advancements in Models and Algorithms for Management Science · MIT · PhD
Tried and failed
unsupervised clustering for symbolic concept discovery applied to neuro-symbolic perception learning. Outcome: no signal. Reason: inability to map learned clusters to meaningful symbolic concepts and scale to complex perception
Neuro-symbolic learning of answer set programs from raw data · Imperial · PhD
Tried and failed
unsupervised Latent Dirichlet Allocation applied to text classification against predefined taxonomy. Reason: unsupervised topic clusters did not correspond one-to-one with predefined target taxonomy categories
Analytics-Enabled Quality and Safety Management Methods for High-Stakes Manufacturing Applications · MIT · PhD
Tried and failed
Unsupervised topic modeling applied to public comments and reviews text. Outcome: no signal. Reason: Failed to produce theoretically or socially meaningful clusters for downstream analysis
Natural Language Processing and Deep Learning Approaches for Sustainability and Infrastructure Policy Analyses · Georgia Tech · PhD
Considered and rejected
Considered and rejected: Rejected unsupervised clustering techniques for project grouping because they do not reflect the relationship between input features and target performance metrics.
Novel approaches to benchmark capital project performance : an application to healthcare projects · UT Austin · PhD
Considered and rejected
Considered and rejected: Rejected unsupervised clustering methods due to insufficient analyst control over high thematic detail category definitions
Generation of a Land Cover Atlas of environmental critic zones using unconventional tools · Politecnico di Torino · PhD
Considered and rejected
Considered and rejected: Decided against using unsupervised clustering methods as the primary behavioral classification platform due to difficulty in tuning hyperparameters, lack of ground truth, and opacity of black-box models.
Rage against the machine: advancing aggression ethology through machine learning · University of Washington · PhD
Considered and rejected
Considered and rejected: Rejected fully automated unsupervised clustering for defining population-level song types due to high-dimensional metric space properties, choosing semi-supervised clustering with manual human validation.
Cultural evolution in the wild: tracking the landscape of learning in bird song · Oxford · PhD
Considered and rejected
Considered and rejected: Rejected unsupervised disentangled representation learning without inductive bias/speaker identity because learned representations can be uninterpretable or meaningless.
Exploring Knowledge Transfer with Deep Learning · Duke · PhD
Considered and rejected
Considered and rejected: Rejected K-means and Learning Vector Quantization (LVQ) clustering algorithms for unsupervised spectral classification because they required more manual effort and did not automatically select the optimal number of classes.
Methods for quantifying changes to Lesser Prairie-Chicken lek connectivity and habitat · Texas Tech · Master's
Supervised machine learning models are beaten by simpler baselines, manual heuristics, or unsupervised methods
Complex supervised classifiers and regression models were outperformed by simple dataset means, single handcrafted statistical tests, or manual annotations of clustering outputs. Supervised methods also yielded higher error rates than unsupervised K-means when handling dominant network attack scenarios, short duration basic feature events, or bridge inspection images.
Lost to a baseline
Mean of training data out-performed all supervised ML models (Ridge, Lasso, ExtraTrees, KernelRidge) across most datasets on held-out perturbation log fold change MSE and MAE.
ASSESSING RELIABILITY OF CAUSAL MODELS OF TRANSCRIPTION · Johns Hopkins · PhD
Lost to a baseline
Unsupervised clustering had a lower MAE (6.45 min) than supervised classification with misclassification (13.8 min) on short duration basic feature events.
Machine learning framework for end to end implementation of incident duration prediction · Iowa State · Master's
Lost to a baseline
Single-best-score cell-type mapping from RCTD decomposed Stereo-Seq bins performed worse than manual annotation of unsupervised Louvain clustering for discerning rare and nuanced CCS populations.
An investigation of the relationship between transcriptomic signals and functional phenotypes across atrial physiology and pharmacology · Oxford · PhD
Lost to a baseline
A 1D chi-square test on the handcrafted invariant mass feature outperformed all supervised multivariate ML classifiers (ANN, BDT, SVM) in discovery power for the two-body decay problem.
The Search for Dark Photons at LHCb and Machine Learning in Particle Physics · MIT · PhD
Lost to a baseline
Supervised SVM exhibited higher classification error than unsupervised K-means clustering in largely malicious network scenarios where honest nodes comprised less than 40% of the network.
Identification and mitigation of attacks in trust-based distributed communication networks · Oxford · PhD
Lost to a baseline
DeepLabV3+ (recall 0.256) and ViT (recall 0.027-0.321) lost to unsupervised K-means (recall 0.777) on the full UAV-collected dataset
Enhancing Structural Inspections by Integrating Computer Vision, Machine Learning, and Risk Assessment for Comprehensive Corrosion Evaluation · Georgia Tech · PhD
Unsupervised methods underperform supervised baselines when labeled ground truth is available
Across pose estimation, sound clustering, energy disaggregation, and video classification, unsupervised models consistently trailed behind their supervised counterparts. Supervised baselines such as random forests and factorial hidden Markov models achieved substantially higher classification accuracy and estimation quality than unsupervised clustering approaches.
Lost to a baseline
On BBCPose supervised baselines, Pfister et al. (88.01% avg accuracy) outperformed the thesis's unsupervised 75.93%
Unsupervised landmark discovery via self-training correspondence · Nottingham · PhD
Lost to a baseline
K-means unsupervised clustering underperformed supervised Random Forest, achieving only 38.2% to 82.4% accuracy compared to RF's 81.6% to 94.4%.
IDENTIFYING MIGRATION FATE AND FACTORS CONTRIBUTING TO MORTALITY OF ATLANTIC SALMON SMOLTS · Dalhousie · Master's
Lost to a baseline
DUE unsupervised disaggregation had lower Estimation Accuracy than supervised Factorial Hidden Markov Models (FHMM)
Flexibility for large-scale deployment of PV systems in low-voltage grids · EPFL · PhD
Lost to a baseline
On Kinetics-Sound clustering, SeLaVi achieved 50.2% NMI and 43.2% accuracy, losing to the supervised baseline which achieved 81.7% NMI and 75.0% accuracy
Learning deep neural networks: necessity and scope of prior knowledge, raw data, and labels · Oxford · PhD
Lost to a baseline
Handcrafted/supervised OmniSource Slow-8x8-R101x2 beat thesis's linear transformer-pooling feature representation on HMDB-51 (83.8% vs 81.3%) and UCF-101 (98.6% vs 98.0%)
Learning and interpreting deep representations from multi-modal data · Oxford · PhD
Advanced unsupervised pipelines and representations are beaten by simpler clustering or raw data baselines
Complex unsupervised pipelines such as autoencoders combined with K-means, Gaussian mixture models, and unsupervised curve alignment failed to outperform raw data clustering or unaligned curves. Additionally, automatic cluster selection criteria like the Bayesian information criterion frequently picked overly small cluster counts that underperformed fixed baselines or alternative clusterers.
Lost to a baseline
Unsupervised alignment methods (unsupervised beta-cdf and registr) failed to outperform unaligned modularity/participation curves in predicting executive function in the PNC cohort.
Statistical and Machine Learning Methods for Neuroimaging and Neurocognitive Data · Penn · PhD
Lost to a baseline
Unsupervised AE+k-means achieved a lower F-statistic (12.4) for heart failure risk separation than k-means applied directly to raw input data (13.2)
Outcome-driven deep clustering for cardiovascular subtyping · Boston University · PhD
Tried and failed
cascade k-means clustering for unsupervised classification applied to remote sensing imagery classification. Outcome: worse than baseline. Reason: underperformed compared to x-means in identifying target spectral clusters accurately
Methods for quantifying changes to Lesser Prairie-Chicken lek connectivity and habitat · Texas Tech · Master's
Lost to a baseline
Unsupervised BIC-driven cluster selection frequently selected K=1 or K=2, performing worse than fixing cluster count a priori to K=4.
STATISTICAL METHODS FOR IDENTIFYING AGING-RELATED VULNERABILITY STATES · Johns Hopkins · PhD
Lost to a baseline
Gaussian Mixture Models (recall 0.239) lost to unsupervised K-means (recall 0.777) on UAV bridge inspection imagery
Enhancing Structural Inspections by Integrating Computer Vision, Machine Learning, and Risk Assessment for Comprehensive Corrosion Evaluation · Georgia Tech · PhD
Supervised models fail due to label noise, subjective rating variance, optimization penalties, or unseen domains
Supervised classifiers and regression models failed when ground-truth labels were noisy during network attacks or when subjective human inspection ratings introduced extreme variability. Furthermore, excessive penalty weighting in risk-aware losses diverted optimization from the primary supervised signal, and models trained on specific attack patterns failed to generalize to unseen strategies.
Tried and failed
risk-aware loss penalty weighting applied to clinical decision support recommendations. Outcome: worse than baseline. Reason: excessive penalty weight diverted the optimization focus away from the primary supervised learning signal
Tried and failed
supervised classification using unmodified benchmark dataset labels applied to network intrusion detection. Reason: ground-truth labels were noisy because malicious agents exhibited normal message intervals during attacks
RSU-Based Intrusion Detection and Autonomous Intersection Response Systems · Virginia Tech · Master's
Considered and rejected
Considered and rejected: Rejected standard supervised classifiers trained on specific cyberattack models because they fail to generalize to unseen attack strategies.
Computational Methods for Fast and Secure Distributed Optimal Power Flow · Georgia Tech · PhD
Tried and failed
direct supervised regression of subjective condition ratings applied to infrastructure asset condition assessment. Outcome: worse than baseline. Reason: models yielded negative R-squared values due to high human inspection variability masking underlying signal
Human Inspection Variability in Infrastructure Asset Management: A Focus on HVAC Systems · Virginia Tech · Master's
Left open by the authors
Problems the authors named and did not get to.
Left open
Develop an unsupervised attribute discovery method to learn articulatory speech features directly from audio without predefined phonetic mappings. Blocker: None
Modeling of Language-Universal Speech Attributes for Multilingual Speech Recognition and Processing · Georgia Tech · PhD
Left open
Develop an unsupervised algorithm to induce mode hierarchy and rāga specifications from unannotated performance audio or symbolic music data. Blocker: None
The Structure of Free Polyphony · EPFL · PhD
Left open
Implement and evaluate supervised machine learning models like random forests to improve the robustness of bioprocess dynamic phase detection. Blocker: Requires dynamic bioprocess time-series run data from the unit operations.
Statistical approaches supporting QbD milestones via bioprocess digital twins · TU Wien · PhD
Left open
Synthesize network classification rules for mixed traffic types using unsupervised clustering or weak rule ensembles. Blocker: None
Practical Network Programming Automation · Penn · PhD
Who has worked on this
Authors of the public theses cited on this page, most-cited first.
- Hana Nicole Herndon · Georgia Tech · PhD · 2026
Enhancing Structural Inspections by Integrating Computer Vision, Machine Learning, and Risk Assessment for Comprehensive Corrosion Evaluation - Justin Dawsey · Texas Tech · Master's · 2025
Methods for quantifying changes to Lesser Prairie-Chicken lek connectivity and habitat - Skylar Jean Lama · Georgia Tech · PhD · 2026
Understanding the Physical Environment of a French Polynesian Coral Reef Ecosystem Using Ocean Modeling, Acoustics, And Machine Learning - Reese Wilson · Texas Tech · PhD · 2026
Liver Abscess Effects on Rumen Histology, Morphology, and Feeding Behavior in BeefÍDairy Cattle, Surveillance of Salmonella Enterica Throughout the Beef Carcass, and Characterization of Salmonella Ent - Charlotte Clapham · Johns Hopkins · PhD · 2026
STATISTICAL METHODS FOR IDENTIFYING AGING-RELATED VULNERABILITY STATES - Rachel July Harris · Georgia Tech · PhD · 2025
Computational Methods for Fast and Secure Distributed Optimal Power Flow - Hao Yen · Georgia Tech · PhD · 2025
Modeling of Language-Universal Speech Attributes for Multilingual Speech Recognition and Processing - Andy Zhao · Cornell · PhD · 2025
Checking, Moderating, Adapting: Collective Engagement for Digital Integrity - Eric M Kernfeld · Johns Hopkins · PhD · 2025
ASSESSING RELIABILITY OF CAUSAL MODELS OF TRANSCRIPTION - Patricia Emogene Miller · Boston University · PhD · 2025
Outcome-driven deep clustering for cardiovascular subtyping - Yishan Shen · Penn · PhD · 2025
Enhancing the Reliability of Real-World Evidence and Clinical Decision-Making: Robust, Calibrated, and Uncertainty-Aware Methods for Observational Healthcare Research - Dawn Parry · Cornell · PhD · 2024
Passive Acoustic Monitoring of Whales and Ocean Ambient Sound to Inform Management
Checking a claim in this area?
We can run the same search on any method or claim. If nothing turns up, we will say so, and that proves nothing on its own.