Recently, self-supervised learning methods based on masked latent prediction have proven to encode input data into powerful representations. However, during training, the learned latent space can be further transformed to extract higher-level information that could be more suited for downstream classification tasks. Therefore, we propose a new method: MAsked latenT Prediction And Classification (MATPAC), which is trained with two pretext tasks solved jointly. As in previous work, the first pretext task is a masked latent prediction task, ensuring a robust input representation in the latent space. The second one is unsupervised classification, which utilises the latent representations of the first pretext task to match probability distributions between a teacher and a student. We validate the MATPAC method by comparing it to other state-of-the-art proposals and conducting ablations studies. MATPAC reaches state-of-the-art self-supervised learning results on reference audio classification datasets such as OpenMIC, GTZAN, ESC-50 and US8K and outperforms comparable supervised methods results for musical auto-tagging on Magna-tag-a-tune.
Multi-modal models, such as CLIP, have demonstrated strong performance in aligning visual and textual representations, excelling in tasks like image retrieval and zero-shot classification. Despite this success, the mechanisms by which these models utilize training data, particularly the role of memorization, remain unclear. In uni-modal models, both supervised and self-supervised, memorization has been shown to be essential for generalization. However, it is not well understood how these findings would apply to CLIP, which incorporates elements from both supervised learning via captions that provide a supervisory signal similar to labels, and from self-supervised learning via the contrastive objective. To bridge this gap in understanding, we propose a formal definition of memorization in CLIP (CLIPMem) and use it to quantify memorization in CLIP models. Our results indicate that CLIP's memorization behavior falls between the supervised and self-supervised paradigms, with "mis-captioned" samples exhibiting highest levels of memorization. Additionally, we find that the text encoder contributes more to memorization than the image encoder, suggesting that mitigation strategies should focus on the text domain. Building on these insights, we propose multiple strategies to reduce memorization while at the same time improving utility--something that had not been shown before for traditional learning paradigms where reducing memorization typically results in utility decrease.
Accurate segmentation of all pathological findings in 3D medical images remains a significant challenge, as supervised models are limited to detecting only the few pathology classes annotated in existing datasets. To address this, we frame pathology segmentation as an unsupervised visual anomaly segmentation (UVAS) problem, leveraging the inherent rarity of pathological patterns compared to healthy ones. We enhance the existing density-based UVAS framework with two key innovations: (1) dense self-supervised learning (SSL) for feature extraction, eliminating the need for supervised pre-training, and (2) learned, masking-invariant dense features as conditioning variables, replacing hand-crafted positional encodings. Trained on over 30,000 unlabeled 3D CT volumes, our model, Screener, outperforms existing UVAS methods on four large-scale test datasets comprising 1,820 scans with diverse pathologies. Code and pre-trained models will be made publicly available.
Humans develop world models that capture the underlying generation process of data. Whether neural networks can learn similar world models remains an open problem. In this work, we provide the first theoretical results for this problem, showing that in a multi-task setting, models with a low-degree bias provably recover latent data-generating variables under mild assumptions -- even if proxy tasks involve complex, non-linear functions of the latents. However, such recovery is also sensitive to model architecture. Our analysis leverages Boolean models of task solutions via the Fourier-Walsh transform and introduces new techniques for analyzing invertible Boolean transforms, which may be of independent interest. We illustrate the algorithmic implications of our results and connect them to related research areas, including self-supervised learning, out-of-distribution generalization, and the linear representation hypothesis in large language models.
Self-supervised learning (SSL) allows training data representations without a supervised signal and has become an important paradigm in machine learning. Most SSL methods employ the cosine similarity between embedding vectors and hence effectively embed data on a hypersphere. While this seemingly implies that embedding norms cannot play any role in SSL, a few recent works have suggested that embedding norms have properties related to network convergence and confidence. In this paper, we resolve this apparent contradiction and systematically establish the embedding norm's role in SSL training. Using theoretical analysis, simulations, and experiments, we show that embedding norms (i) govern SSL convergence rates and (ii) encode network confidence, with smaller norms corresponding to unexpected samples. Additionally, we show that manipulating embedding norms can have large effects on convergence speed. Our findings demonstrate that SSL embedding norms are integral to understanding and optimizing network behavior.
In remote healthcare monitoring, time series representation learning reveals critical patient behavior patterns from high-frequency data. This study analyzes home activity data from individuals living with dementia by proposing a two-stage, self-supervised learning approach tailored to uncover low-rank structures. The first stage converts time-series activities into text sequences encoded by a pre-trained language model, providing a rich, high-dimensional latent state space using a PageRank-based method. This PageRank vector captures latent state transitions, effectively compressing complex behaviour data into a succinct form that enhances interpretability. This low-rank representation not only enhances model interpretability but also facilitates clustering and transition analysis, revealing key behavioral patterns correlated with clinicalmetrics such as MMSE and ADAS-COG scores. Our findings demonstrate the framework's potential in supporting cognitive status prediction, personalized care interventions, and large-scale health monitoring.
Online Continual Learning (OCL) methods train a model on a non-stationary data stream where only a few examples are available at a time, often leveraging replay strategies. However, usage of replay is sometimes forbidden, especially in applications with strict privacy regulations. Therefore, we propose Continual MultiPatches (CMP), an effective plug-in for existing OCL self-supervised learning strategies that avoids the use of replay samples. CMP generates multiple patches from a single example and projects them into a shared feature space, where patches coming from the same example are pushed together without collapsing into a single point. CMP surpasses replay and other SSL-based strategies on OCL streams, challenging the role of replay as a go-to solution for self-supervised OCL.
Pretraining models with unsupervised graph representation learning has led to significant advancements in domains such as social network analysis, molecular design, and electronic design automation (EDA). However, prior work in EDA has mainly focused on pretraining models for digital circuits, overlooking analog and mixed-signal circuits. To bridge this gap, we introduce DICE, a Device-level Integrated Circuits Encoder, which is the first graph neural network (GNN) pretrained via self-supervised learning specifically tailored for graph-level prediction tasks in both analog and digital circuits. DICE adopts a simulation-free pretraining approach based on graph contrastive learning, leveraging two novel graph augmentation techniques. Experimental results demonstrate substantial performance improvements across three downstream tasks, highlighting the effectiveness of DICE for both analog and digital circuits. The code is available at github.com/brianlsy98/DICE.
From crop mapping to flood detection, machine learning in remote sensing has a wide range of societally beneficial applications. The commonalities between remote sensing data in these applications present an opportunity for pretrained machine learning models tailored to remote sensing to reduce the labeled data and effort required to solve individual tasks. However, such models must be: (i) flexible enough to ingest input data of varying sensor modalities and shapes (i.e., of varying spatial and temporal dimensions), and (ii) able to model Earth surface phenomena of varying scales and types. To solve this gap, we present Galileo, a family of pretrained remote sensing models designed to flexibly process multimodal remote sensing data. We also introduce a novel and highly effective self-supervised learning approach to learn both large- and small-scale features, a challenge not addressed by previous models. Our Galileo models obtain state-of-the-art results across diverse remote sensing tasks.
Automated lesion segmentation of medical images has made tremendous improvements in recent years due to deep learning advancements. However, accurately capturing fine-grained global and regional feature representations remains a challenge. Many existing methods obtain suboptimal performance on complex lesion segmentation due to information loss during typical downsampling operations and the insufficient capture of either regional or global features. To address these issues, we propose the Global and Regional Compensation Segmentation Framework (GRCSF), which introduces two key innovations: the Global Compensation Unit (GCU) and the Region Compensation Unit (RCU). The proposed GCU addresses resolution loss in the U-shaped backbone by preserving global contextual features and fine-grained details during multiscale downsampling. Meanwhile, the RCU introduces a self-supervised learning (SSL) residual map generated by Masked Autoencoders (MAE), obtained as pixel-wise differences between reconstructed and original images, to highlight regions with potential lesions. These SSL residual maps guide precise lesion localization and segmentation through a patch-based cross-attention mechanism that integrates regional spatial and pixel-level features. Additionally, the RCU incorporates patch-level importance scoring to enhance feature fusion by leveraging global spatial information from the backbone. Experiments on two publicly available medical image segmentation datasets, including brain stroke lesion and coronary artery calcification datasets, demonstrate that our GRCSF outperforms state-of-the-art methods, confirming its effectiveness across diverse lesion types and its potential as a generalizable lesion segmentation solution.
In recent years, the field of autonomous driving has attracted increasingly significant public interest. Accurately forecasting the future behavior of various traffic participants is essential for the decision-making of Autonomous Vehicles (AVs). In this paper, we focus on both scenario-based and perception-based motion forecasting for AVs. We propose a formal problem formulation for motion forecasting and summarize the main challenges confronting this area of research. We also detail representative datasets and evaluation metrics pertinent to this field. Furthermore, this study classifies recent research into two main categories: supervised learning and self-supervised learning, reflecting the evolving paradigms in both scenario-based and perception-based motion forecasting. In the context of supervised learning, we thoroughly examine and analyze each key element of the methodology. For self-supervised learning, we summarize commonly adopted techniques. The paper concludes and discusses potential research directions, aiming to propel progress in this vital area of AV technology.
Employing self-supervised learning (SSL) methodologies assumes par-amount significance in handling unlabeled polyp datasets when building deep learning-based automatic polyp segmentation models. However, the intricate privacy dynamics surrounding medical data often preclude seamless data sharing among disparate medical centers. Federated learning (FL) emerges as a formidable solution to this privacy conundrum, yet within the realm of FL, optimizing model generalization stands as a pressing imperative. Robust generalization capabilities are imperative to ensure the model's efficacy across diverse geographical domains post-training on localized client datasets. In this paper, a Federated self-supervised Domain Generalization method is proposed to enhance the generalization capacity of federated and Label-efficient intestinal polyp segmentation, named LFDG. Based on a classical SSL method, DropPos, LFDG proposes an adversarial learning-based data augmentation method (SSADA) to enhance the data diversity. LFDG further proposes a relaxation module based on Source-reconstruction and Augmentation-masking (SRAM) to maintain stability in feature learning. We have validated LFDG on polyp images from six medical centers. The performance of our method achieves 3.80% and 3.92% better than the baseline and other recent FL methods and SSL methods, respectively.
The nonlinear, non-convex AC Optimal Power Flow (AC-OPF) problem is fundamental for power systems operations. The intrinsic complexity of AC-OPF has fueled a growing interest in the development of optimization proxies for the problem, i.e., machine learning models that predict high-quality, close-to-optimal solutions. More recently, dual conic proxy architectures have been proposed, which combine machine learning and convex relaxations of AC-OPF, to provide valid certificates of optimality using learning-based methods. Building on this methodology, this paper proposes, for the first time, a dual conic proxy architecture for the semidefinite (SDP) relaxation of AC-OPF problems. Although the SDP relaxation is stronger than the second-order cone relaxation considered in previous work, its practical use has been hindered by its computational cost. The proposed method combines a neural network with a differentiable dual completion strategy that leverages the structure of the dual SDP problem. This approach guarantees dual feasibility, and therefore valid dual bounds, while providing orders of magnitude of speedups compared to interior-point algorithms. The paper also leverages self-supervised learning, which alleviates the need for time-consuming data generation and allows to train the proposed models efficiently. Numerical experiments are presented on several power grid benchmarks with up to 500 buses. The results demonstrate that the proposed SDP-based proxies can outperform weaker conic relaxations, while providing several orders of magnitude speedups compared to a state-of-the-art interior-point SDP solver.
Traditionally, Knowledge Distillation (KD) is used for model compression, often leading to suboptimal performance. In this paper, we evaluate the impact of combining KD loss with alternative pruning techniques, including Low-Rank Factorization (LRF) and l0 regularization, on a conformer-based pre-trained network under the paradigm of Self-Supervised Learning (SSL). We also propose a strategy to jointly prune and train an RNN-T-based ASR model, demonstrating that this approach yields superior performance compared to pruning a pre-trained network first and then using it for ASR training. This approach led to a significant reduction in word error rate: l0 and KD combination achieves the best non-streaming performance, with a 8.9% Relative Word Error Rate (RWER) improvement over the baseline, while LRF and KD combination yields the best results for streaming ASR, improving RWER by 13.4%.
Lipreading is an important technique for facilitating human-computer interaction in noisy environments. Our previously developed self-supervised learning method, AV2vec, which leverages multimodal self-distillation, has demonstrated promising performance in speaker-independent lipreading on the English LRS3 dataset. However, AV2vec faces challenges such as high training costs and a potential scarcity of audio-visual data for lipreading in languages other than English, such as Chinese. Additionally, most studies concentrate on speakerindependent lipreading models, which struggle to account for the substantial variation in speaking styles across di?erent speakers. To address these issues, we propose a comprehensive approach. First, we investigate cross-lingual transfer learning, adapting a pre-trained AV2vec model from a source language and optimizing it for the lipreading task in a target language. Second, we enhance the accuracy of lipreading for specific target speakers through a speaker adaptation strategy, which is not extensively explored in previous research. Third, after analyzing the complementary performance of lipreading with lip region-of-interest (ROI) and face inputs, we introduce a model ensembling strategy that integrates both, signi?cantly boosting model performance. Our method achieved a character error rate (CER) of 77.3% on the evaluation set of the ChatCLR dataset, which is lower than the top result from the 2024 Chat-scenario Chinese Lipreading Challenge.
One of the methods for language Identification (LID) involves deriving speech representation from pre-trained models using self-supervised learning, followed by fine-tuning the model for the LID task. State-of-the-art approaches for LID use an attention-based statistical pooling layer to facilitate the aggregation of contextual information across time frames of the embedding vectors extracted from the pre-trained model. In this paper, we delve into exploring recently proposed attention mechanisms, namely performer and agent-attention, in conjunction with the statistical pooling layer. The LID experiments are performed on three datasets: VoxPopuli, FLEURS, and VoxLingua. We compare their performance against vanilla self-attention. Our findings suggest that performer-attention outperforms self-attention and agent-attention exhibits comparable or occasionally superior performance to self-attention, while also being computationally less expensive.
The problem of synthetic speech detection has enjoyed considerable attention, with recent methods achieving low error rates across several established benchmarks. However, to what extent can low error rates on academic benchmarks translate to more realistic conditions? In practice, while the training set is fixed at one point in time, test-time conditions may exhibit distribution shifts relative to the training conditions, such as changes in speaker characteristics, emotional expressiveness, language and acoustic conditions, and the emergence of novel synthesis methods. Although some existing datasets target subsets of these distribution shifts, systematic analysis remains difficult due to inconsistencies between source data and synthesis systems across datasets. This difficulty is further exacerbated by the rapid development of new text-to-speech (TTS) and vocoder systems, which continually expand the diversity of synthetic speech. To enable systematic benchmarking of model performance under distribution shifts, we introduce ShiftySpeech, a large-scale benchmark comprising over 3,000 hours of synthetic speech across 7 source domains, 6 TTS systems, 12 vocoders, and 3 languages. ShiftySpeech is specifically designed to evaluate model generalization under controlled distribution shifts while ensuring broad coverage of modern synthetic speech generation techniques. It fills a key gap in current benchmarks by supporting fine-grained, controlled analysis of generalization robustness. All tested distribution shifts significantly degrade detection performance of state-of-the-art detection approaches based on self-supervised features. Overall, our findings suggest that reliance on synthetic speech detection methods in production environments should be carefully evaluated based on anticipated distribution shifts.
The Multi-Agent Path Finding (MAPF) problem aims to determine the shortest and collision-free paths for multiple agents in a known, potentially obstacle-ridden environment. It is the core challenge for robotic deployments in large-scale logistics and transportation. Decentralized learning-based approaches have shown great potential for addressing the MAPF problems, offering more reactive and scalable solutions. However, existing learning-based MAPF methods usually rely on agents making decisions based on a limited field of view (FOV), resulting in short-sighted policies and inefficient cooperation in complex scenarios. There, a critical challenge is to achieve consensus on potential movements between agents based on limited observations and communications. To tackle this challenge, we introduce a new framework that applies sheaf theory to decentralized deep reinforcement learning, enabling agents to learn geometric cross-dependencies between each other through local consensus and utilize them for tightly cooperative decision-making. In particular, sheaf theory provides a mathematical proof of conditions for achieving global consensus through local observation. Inspired by this, we incorporate a neural network to approximately model the consensus in latent space based on sheaf theory and train it through self-supervised learning. During the task, in addition to normal features for MAPF as in previous works, each agent distributedly reasons about a learned consensus feature, leading to efficient cooperation on pathfinding and collision avoidance. As a result, our proposed method demonstrates significant improvements over state-of-the-art learning-based MAPF planners, especially in relatively large and complex scenarios, demonstrating its superiority over baselines in various simulations and real-world robot experiments.
The proliferation of IoT devices has significantly increased network vulnerabilities, creating an urgent need for effective Intrusion Detection Systems (IDS). Machine Learning-based IDS (ML-IDS) offer advanced detection capabilities but rely on labeled attack data, which limits their ability to identify unknown threats. Self-Supervised Learning (SSL) presents a promising solution by using only normal data to detect patterns and anomalies. This paper introduces SAFE, a novel framework that transforms tabular network intrusion data into an image-like format, enabling Masked Autoencoders (MAEs) to learn robust representations of network behavior. The features extracted by the MAEs are then incorporated into a lightweight novelty detector, enhancing the effectiveness of anomaly detection. Experimental results demonstrate that SAFE outperforms the state-of-the-art anomaly detection method, Scale Learning-based Deep Anomaly Detection method (SLAD), by up to 26.2% and surpasses the state-of-the-art SSL-based network intrusion detection approach, Anomal-E, by up to 23.5% in F1-score.
The effectiveness of neural network models largely relies on learning meaningful latent patterns from data, where self-supervised learning of informative representations can enhance model performance and generalisability. However, self-supervised representation learning for spatially characterised time series, which are ubiquitous in transportation domain, poses unique challenges due to the necessity of maintaining fine-grained spatio-temporal similarities in the latent space. In this study, we introduce two structure-preserving regularisers for the contrastive learning of spatial time series: one regulariser preserves the topology of similarities between instances, and the other preserves the graph geometry of similarities across spatial and temporal dimensions. To balance the contrastive learning objective and the need for structure preservation, we propose a dynamic weighting mechanism that adaptively manages this trade-off and stabilises training. We validate the proposed method through extensive experiments, including multivariate time series classification to demonstrate its general applicability, as well as macroscopic and microscopic traffic prediction to highlight its particular usefulness in encoding traffic interactions. Across all tasks, our method preserves the similarity structures more effectively and improves state-of-the-art task performances. This method can be integrated with an arbitrary neural network model and is particularly beneficial for time series data with spatial or geographical features. Furthermore, our findings suggest that well-preserved similarity structures in the latent space indicate more informative and useful representations. This provides insights to design more effective neural networks for data-driven transportation research. Our code is made openly accessible with all resulting data at https://github.com/yiru-jiao/spclt