Self-supervised learning (SSL) for speech representation has been successfully applied in various downstream tasks, such as speech and speaker recognition. More recently, speech SSL models have also been shown to be beneficial in advancing spoken language understanding tasks, implying that the SSL models have the potential to learn not only acoustic but also linguistic information. In this paper, we aim to clarify if speech SSL techniques can well capture linguistic knowledge. For this purpose, we introduce SpeechGLUE, a speech version of the General Language Understanding Evaluation (GLUE) benchmark. Since GLUE comprises a variety of natural language understanding tasks, SpeechGLUE can elucidate the degree of linguistic ability of speech SSL models. Experiments demonstrate that speech SSL models, although inferior to text-based SSL models, perform better than baselines, suggesting that they can acquire a certain amount of general linguistic knowledge from just unlabeled speech data.
The excellent generalization ability of self-supervised learning (SSL) for speech foundation models has garnered significant attention. HuBERT is a successful example that utilizes offline clustering to convert speech features into discrete units for a masked language modeling pretext task. However, simply clustering features as targets by k-means does not fully inspire the model's performance. In this work, we present an unsupervised method to improve SSL targets. Two models are proposed, MonoBERT and PolyBERT, which leverage context-independent and context-dependent phoneme-based units for pre-training. Our models outperform other SSL models significantly on the LibriSpeech benchmark without the need for iterative re-clustering and re-training. Furthermore, our models equipped with context-dependent units even outperform target-improvement models that use labeled data during pre-training. How we progressively improve the unit discovery process is demonstrated through experiments.
Text Classification is one of the fundamental tasks in natural language processing, which requires an agent to determine the most appropriate category for input sentences. Recently, deep neural networks have achieved impressive performance in this area, especially Pre-trained Language Models (PLMs). Usually, these methods concentrate on input sentences and corresponding semantic embedding generation. However, for another essential component: labels, most existing works either treat them as meaningless one-hot vectors or use vanilla embedding methods to learn label representations along with model training, underestimating the semantic information and guidance that these labels reveal. To alleviate this problem and better exploit label information, in this paper, we employ Self-Supervised Learning (SSL) in model learning process and design a novel self-supervised Relation of Relation (R2) classification task for label utilization from a one-hot manner perspective. Then, we propose a novel Relation of Relation Learning Network (R2-Net) for text classification, in which text classification and R2 classification are treated as optimization targets. Meanwhile, triplet loss is employed to enhance the analysis of differences and connections among labels. Moreover, considering that one-hot usage is still short of exploiting label information, we incorporate external knowledge from WordNet to obtain multi-aspect descriptions for label semantic learning and extend R2-Net to a novel Description-Enhanced Label Embedding network (DELE) from a label embedding perspective. ...
Self-supervised learning has shown its promising capability in graph representation learning in recent work. Most existing pre-training strategies usually choose the popular Graph neural networks (GNNs), which can be seen as a special form of low-pass filter, fail to effectively capture heterophily. In this paper, we first present an experimental investigation exploring the performance of low-pass and high-pass filters in heterophily graph classification, where the results clearly show that high-frequency signal is important for learning heterophily graph representation. On the other hand, it is still unclear how to effectively capture the structural pattern of graphs and how to measure the capability of the self-supervised pre-training strategy in capturing graph structure. To address the problem, we first design a quantitative metric to Measure Graph Structure (MGS), which analyzes correlation between structural similarity and embedding similarity of graph pairs. Then, to enhance the graph structural information captured by self-supervised learning, we propose a novel self-supervised strategy for Pre-training GNNs based on the Metric (PGM). Extensive experiments validate our pre-training strategy achieves state-of-the-art performance for molecular property prediction and protein function prediction. In addition, we find choosing the suitable filter sometimes may be better than designing good pre-training strategies for heterophily graph classification.
Large, pre-trained representation models trained using self-supervised learning have gained popularity in various fields of machine learning because they are able to extract high-quality salient features from input data. As such, they have been frequently used as base networks for various pattern classification tasks such as speech recognition. However, not much research has been conducted on applying these types of models to the field of speech signal generation. In this paper, we investigate the feasibility of using pre-trained speech representation models for a downstream speech enhancement task. To alleviate mismatches between the input features of the pre-trained model and the target enhancement model, we adopt a novel feature normalization technique to smoothly link these modules together. Our proposed method enables significant improvements in speech quality compared to baselines when combined with various types of pre-trained speech models.
Deep learning is the state-of-the-art for medical imaging tasks, but requires large, labeled datasets. For risk prediction, large datasets are rare since they require both imaging and follow-up (e.g., diagnosis codes). However, the release of publicly available imaging data with diagnostic labels presents an opportunity for self and semi-supervised approaches to improve label efficiency for risk prediction. Though several studies have compared self-supervised approaches in natural image classification, object detection, and medical image interpretation, there is limited data on which approaches learn robust representations for risk prediction. We present a comparison of semi- and self-supervised learning to predict mortality risk using chest x-ray images. We find that a semi-supervised autoencoder outperforms contrastive and transfer learning in internal and external validation.
Finding multiple temporal relationships among locations can benefit a bunch of urban applications, such as dynamic offline advertising and smart public transport planning. While some efforts have been made on finding static relationships among locations, little attention is focused on studying time-aware location relationships. Indeed, abundant location-based human activities are time-varying and the availability of these data enables a new paradigm for understanding the dynamic relationships in a period among connective locations. To this end, we propose to study a new problem, namely multi-Temporal relationship inference among locations (Trial for short), where the major challenge is how to integrate dynamic and geographical influence under the relationship sparsity constraint. Specifically, we propose a solution to Trial with a graph learning scheme, which includes a spatially evolving graph neural network (SEENet) with two collaborative components: spatially evolving graph convolution module (SEConv) and spatially evolving self-supervised learning strategy (SE-SSL). SEConv performs the intra-time aggregation and inter-time propagation to capture the multifaceted spatially evolving contexts from the view of location message passing. In addition, SE-SSL designs time-aware self-supervised learning tasks in a global-local manner with additional evolving constraint to enhance the location representation learning and further handle the relationship sparsity. Finally, experiments on four real-world datasets demonstrate the superiority of our method over several state-of-the-art approaches.
Contrastive cross-modality pretraining has recently exhibited impressive success in diverse fields, whereas there is limited research on their merits in speech emotion recognition (SER). In this paper, we propose GEmo-CLAP, a kind of gender-attribute-enhanced contrastive language-audio pretraining (CLAP) method for SER. Specifically, we first construct an effective emotion CLAP (Emo-CLAP) for SER, using pre-trained text and audio encoders. Second, given the significance of gender information in SER, two novel multi-task learning based GEmo-CLAP (ML-GEmo-CLAP) and soft label based GEmo-CLAP (SL-GEmo-CLAP) models are further proposed to incorporate gender information of speech signals, forming more reasonable objectives. Experiments on IEMOCAP indicate that our proposed two GEmo-CLAPs consistently outperform Emo-CLAP with different pre-trained models. Remarkably, the proposed WavLM-based SL-GEmo-CLAP obtains the best WAR of 83.16\%, which performs better than state-of-the-art SER methods.
The representation degeneration problem is a phenomenon that is widely observed among self-supervised learning methods based on Transformers. In NLP, it takes the form of anisotropy, a singular property of hidden representations which makes them unexpectedly close to each other in terms of angular distance (cosine-similarity). Some recent works tend to show that anisotropy is a consequence of optimizing the cross-entropy loss on long-tailed distributions of tokens. We show in this paper that anisotropy can also be observed empirically in language models with specific objectives that should not suffer directly from the same consequences. We also show that the anisotropy problem extends to Transformers trained on other modalities. Our observations tend to demonstrate that anisotropy might actually be inherent to Transformers-based models.
Simultaneous Localization and Mapping (SLAM) stands as one of the critical challenges in robot navigation. A SLAM system often consists of a front-end component for motion estimation and a back-end system for eliminating estimation drifts. Recent advancements suggest that data-driven methods are highly effective for front-end tasks, while geometry-based methods continue to be essential in the back-end processes. However, such a decoupled paradigm between the data-driven front-end and geometry-based back-end can lead to sub-optimal performance, consequently reducing the system's capabilities and generalization potential. To solve this problem, we proposed a novel self-supervised imperative learning framework, named imperative SLAM (iSLAM), which fosters reciprocal correction between the front-end and back-end, thus enhancing performance without necessitating any external supervision. Specifically, we formulate the SLAM problem as a bilevel optimization so that the front-end and back-end are bidirectionally connected. As a result, the front-end model can learn global geometric knowledge obtained through pose graph optimization by back-propagating the residuals from the back-end component. We showcase the effectiveness of this new framework through an application of stereo-inertial SLAM. The experiments show that the iSLAM training strategy achieves an accuracy improvement of 22% on average over a baseline model. To the best of our knowledge, iSLAM is the first SLAM system showing that the front-end and back-end components can mutually correct each other in a self-supervised manner.
Unlike existing fully-supervised approaches, we rethink colorectal polyp segmentation from an out-of-distribution perspective with a simple but effective self-supervised learning approach. We leverage the ability of masked autoencoders -- self-supervised vision transformers trained on a reconstruction task -- to learn in-distribution representations; here, the distribution of healthy colon images. We then perform out-of-distribution reconstruction and inference, with feature space standardisation to align the latent distribution of the diverse abnormal samples with the statistics of the healthy samples. We generate per-pixel anomaly scores for each image by calculating the difference between the input and reconstructed images and use this signal for out-of-distribution (ie, polyp) segmentation. Experimental results on six benchmarks show that our model has excellent segmentation performance and generalises across datasets. Our code is publicly available at https://github.com/GewelsJI/Polyp-OOD.
We present Malafide, a universal adversarial attack against automatic speaker verification (ASV) spoofing countermeasures (CMs). By introducing convolutional noise using an optimised linear time-invariant filter, Malafide attacks can be used to compromise CM reliability while preserving other speech attributes such as quality and the speaker's voice. In contrast to other adversarial attacks proposed recently, Malafide filters are optimised independently of the input utterance and duration, are tuned instead to the underlying spoofing attack, and require the optimisation of only a small number of filter coefficients. Even so, they degrade CM performance estimates by an order of magnitude, even in black-box settings, and can also be configured to overcome integrated CM and ASV subsystems. Integrated solutions that use self-supervised learning CMs, however, are more robust, under both black-box and white-box settings.
Self-supervised learning models have been shown to learn rich visual representations without requiring human annotations. However, in many real-world scenarios, labels are partially available, motivating a recent line of work on semi-supervised methods inspired by self-supervised principles. In this paper, we propose a conceptually simple yet empirically powerful approach to turn clustering-based self-supervised methods such as SwAV or DINO into semi-supervised learners. More precisely, we introduce a multi-task framework merging a supervised objective using ground-truth labels and a self-supervised objective relying on clustering assignments with a single cross-entropy loss. This approach may be interpreted as imposing the cluster centroids to be class prototypes. Despite its simplicity, we provide empirical evidence that our approach is highly effective and achieves state-of-the-art performance on CIFAR100 and ImageNet.
Self-supervised learning on graphs has made large strides in achieving great performance in various downstream tasks. However, many state-of-the-art methods suffer from a number of impediments, which prevent them from realizing their full potential. For instance, contrastive methods typically require negative sampling, which is often computationally costly. While non-contrastive methods avoid this expensive step, most existing methods either rely on overly complex architectures or dataset-specific augmentations. In this paper, we ask: Can we borrow from classical unsupervised machine learning literature in order to overcome those obstacles? Guided by our key insight that the goal of distance-based clustering closely resembles that of contrastive learning: both attempt to pull representations of similar items together and dissimilar items apart. As a result, we propose CARL-G - a novel clustering-based framework for graph representation learning that uses a loss inspired by Cluster Validation Indices (CVIs), i.e., internal measures of cluster quality (no ground truth required). CARL-G is adaptable to different clustering methods and CVIs, and we show that with the right choice of clustering method and CVI, CARL-G outperforms node classification baselines on 4/5 datasets with up to a 79x training speedup compared to the best-performing baseline. CARL-G also performs at par or better than baselines in node clustering and similarity search tasks, training up to 1,500x faster than the best-performing baseline. Finally, we also provide theoretical foundations for the use of CVI-inspired losses in graph representation learning.
In this work, we investigate the time series representation learning problem using self-supervised techniques. Contrastive learning is well-known in this area as it is a powerful method for extracting information from the series and generating task-appropriate representations. Despite its proficiency in capturing time series characteristics, these techniques often overlook a critical factor - the inherent noise in this type of data, a consideration usually emphasized in general time series analysis. Moreover, there is a notable absence of attention to developing efficient yet lightweight encoder architectures, with an undue focus on delivering contrastive losses. Our work address these gaps by proposing an innovative training strategy that promotes consistent representation learning, accounting for the presence of noise-prone signals in natural time series. Furthermore, we propose an encoder architecture that incorporates dilated convolution within the Inception block, resulting in a scalable and robust network with a wide receptive field. Experimental findings underscore the effectiveness of our method, consistently outperforming state-of-the-art approaches across various tasks, including forecasting, classification, and abnormality detection. Notably, our method attains the top rank in over two-thirds of the classification UCR datasets, utilizing only 40% of the parameters compared to the second-best approach. Our source code for CoInception framework is accessible at https://github.com/anhduy0911/CoInception.
This study is focused on understanding and quantifying the change in phoneme and prosody information encoded in the Self-Supervised Learning (SSL) model, brought by an accent identification (AID) fine-tuning task. This problem is addressed based on model probing. Specifically, we conduct a systematic layer-wise analysis of the representations of the Transformer layers on a phoneme correlation task, and a novel word-level prosody prediction task. We compare the probing performance of the pre-trained and fine-tuned SSL models. Results show that the AID fine-tuning task steers the top 2 layers to learn richer phoneme and prosody representation. These changes share some similarities with the effects of fine-tuning with an Automatic Speech Recognition task. In addition, we observe strong accent-specific phoneme representations in layer 9. To sum up, this study provides insights into the understanding of SSL features and their interactions with fine-tuning tasks.
Limited availability of labeled physiological data often prohibits the use of powerful supervised deep learning models in the biomedical machine intelligence domain. We approach this problem and propose a novel encoding framework that relies on self-supervised learning with momentum contrast to learn representations from multivariate time-series of various physiological domains without needing labels. Our model uses a transformer architecture that can be easily adapted to classification problems by optimizing a linear output classification layer. We experimentally evaluate our framework using two publicly available physiological datasets from different domains, i.e., human activity recognition from embedded inertial sensory and emotion recognition from electroencephalography. We show that our self-supervised learning approach can indeed learn discriminative features which can be exploited in downstream classification tasks. Our work enables the development of domain-agnostic intelligent systems that can effectively analyze multivariate time-series data from physiological domains.
Self-supervised learning (SSL) has led to great strides in speech processing. However, the resources needed to train these models has become prohibitively large as they continue to scale. Currently, only a few groups with substantial resources are capable of creating SSL models, which harms reproducibility. In this work, we optimize HuBERT SSL to fit in academic constraints. We reproduce HuBERT independently from the original implementation, with no performance loss. Our code and training optimizations make SSL feasible with only 8 GPUs, instead of the 32 used in the original work. We also explore a semi-supervised route, using an ASR model to skip the first pre-training iteration. Within one iteration of pre-training, our models improve over HuBERT on several tasks. Furthermore, our HuBERT Large variant requires only 8 GPUs, achieving similar performance to the original trained on 128. As our contribution to the community, all models, configurations, and code are made open-source in ESPnet.
Self-supervised learning (SSL) is a powerful technique for learning from unlabeled data. By learning to remain invariant to applied data augmentations, methods such as SimCLR and MoCo can reach quality on par with supervised approaches. However, this invariance may be detrimental for solving downstream tasks that depend on traits affected by augmentations used during pretraining, such as color. In this paper, we propose to foster sensitivity to such characteristics in the representation space by modifying the projector network, a common component of self-supervised architectures. Specifically, we supplement the projector with information about augmentations applied to images. For the projector to take advantage of this auxiliary conditioning when solving the SSL task, the feature extractor learns to preserve the augmentation information in its representations. Our approach, coined Conditional Augmentation-aware Self-supervised Learning (CASSLE), is directly applicable to typical joint-embedding SSL methods regardless of their objective functions. Moreover, it does not require major changes in the network architecture or prior knowledge of downstream tasks. In addition to an analysis of sensitivity towards different data augmentations, we conduct a series of experiments, which show that CASSLE improves over various SSL methods, reaching state-of-the-art performance in multiple downstream tasks.
In conventional supervised deep learning based channel estimation algorithms, a large number of training samples are required for offline training. However, in practical communication systems, it is difficult to obtain channel samples for every signal-to-noise ratio (SNR). Furthermore, the generalization ability of these deep neural networks (DNN) is typically poor. In this work, we propose a one-shot self-supervised learning framework for channel estimation in multi-input multi-output (MIMO) systems. The required number of samples for offline training is small and our approach can be directly deployed to adapt to variable channels. Our framework consists of a traditional channel estimation module and a denoising module. The denoising module is designed based on the one-shot learning method Self2Self and employs Bernoulli sampling to generate training labels. Besides,we further utilize a blind spot strategy and dropout technique to avoid overfitting. Simulation results show that the performance of the proposed one-shot self-supervised learning method is very close to the supervised learning approach while obtaining improved generalization ability for different channel environments.