This paper introduces a method to select and weight pretext tasks for better self-supervised speech representation learning.
problem Combining pretext tasks for better performance in self-supervised speech representation learning.
method Estimating calibrated weights for partial losses corresponding to pretext tasks during self-supervised training.
result Groups of selected and weighted pretext tasks perform better than classic baselines in automatic speech recognition and speaker/emotion recognition.
Δ-UQ uses anchoring to estimate uncertainty in models.
problem Estimating uncertainty in predictive models.
method Anchoring input into a tuple for uncertainty estimation.
result Δ-UQ outperforms baselines in various use-cases.
The rotation prediction (Rotation) is a simple pretext-task for self-supervised learning (SSL), where models learn useful representations for target vision tasks by solving pretext-tasks. Although Rotation captures information of object shapes, it hardly captures information of textures. To tackle this problem, we intr…
Theoretical analysis shows pretext-based self-supervised learning can be boosted by downstream data under certain conditions.
problem Theoretical analysis of pretext-based self-supervised learning and downstream data refinement.
method Theoretical analysis and experiments on synthetic and real-world datasets.
result Theoretical lower bounds and experiments show that downstream data refinement can boost or hurt performance depending on conditions.
Self-supervised learning improves by predicting known information, reducing labeled data needs.
problem Efficiently learn useful semantic representations without labeled data.
method Develops a mechanism exploiting statistical connections between pretext tasks to learn representations that solve downstream tasks.
result Proves linear layer yields small approximation error and drastically reduces labeled sample complexity.
As a newly emerging unsupervised learning paradigm, self-supervised learning (SSL) recently gained widespread attention, which usually introduces a pretext task without manual annotation of data. With its help, SSL effectively learns the feature representation beneficial for downstream tasks. Thus the pretext task play…
This work improves adaptive conformal prediction using self-supervised learning.
problem Improving the adaptability of conformal prediction intervals.
method Train an auxiliary model with a self-supervised pretext task on top of an existing predictive model and use the self-supervised error as an additional feature to estimate nonconformity scores.
result Empirically demonstrates the benefit of additional information in improving the efficiency (width), deficit, and excess of conformal prediction intervals.
We exploit a self-supervised deep multi-task learning framework for electrocardiogram (ECG) -based emotion recognition. The proposed solution consists of two stages of learning a) learning ECG representations and b) learning to classify emotions. ECG representations are learned by a signal transformation recognition ne…
This paper explores SSL for graph neural networks, improving performance on real-world datasets.
problem Leveraging unlabeled data for graph neural networks to improve deep learning performance.
method Empirical study of various SSL pretext tasks on graphs and proposing a new approach called SelfTask.
result Proposes SelfTask, achieving state-of-the-art performance on real-world datasets.
SuNCEt accelerates contrastive learning with minimal labeled data.
problem Efficiently learning visual representations with limited labeled data.
method Noise-contrastive estimation and neighbourhood component analysis-based semi-supervised loss.
result SuNCEt achieves semi-supervised learning accuracy with less than half the labeled data.
Expanding self-supervised learning to diverse domains reveals Rotation's semantic superiority.
problem Limited self-supervised learning experiments on diverse domains.
method Experimented on various domains (satellite, textural, biological) using popular self-supervised methods.
result Rotation task is semantically most meaningful, with other tasks relying on distribution rather than semantic understanding.
This paper presents a novel yet intuitive approach to unsupervised feature learning. Inspired by the human visual system, we explore whether low-level motion-based grouping cues can be used to learn an effective visual representation. Specifically, we use unsupervised motion-based segmentation on videos to obtain segme…
New unsupervised transfer learning method for spatiotemporal tasks.
problem Transfer knowledge from unsupervised models to new predictive tasks.
method Differentiable framework with Transferable Memory Unit (TMU).
result Significant improvements on spatiotemporal prediction benchmarks.
Novel method reduces radiomic data annotation needs.
problem Insufficient labeled radiomic data for disease diagnosis.
method Collaborative self-supervised learning with two pretext tasks.
result Outperforms other self-supervised methods on radiomic data.
Perception technologies in Autonomous Driving are experiencing their golden age due to the advances in Deep Learning. Yet, most of these systems rely on the semantically rich information of RGB images. Deep Learning solutions applied to the data of other sensors typically mounted on autonomous cars (e.g. lidars or rada…
Unified framework for semi-supervised learning reduces annotation needs.
problem Sparse annotations and large amounts of unlabeled data in computational pathology.
method S5CL integrates fully-supervised, self-supervised, and semi-supervised learning through hierarchical contrastive losses.
result S5CL improves accuracy and F1-score in histopathological datasets with sparse labels.
EQ-Net combines LLR estimation and quantization using deep learning.
problem Unified solution for LLR estimation and quantization.
method Two-stage algorithm using LLR compression as a pretext task.
result Achieves state-of-the-art results with gains in efficiency and latency.
Paper tackles singularity detection in PDEs using data-driven self-supervised learning.
problem Detecting singularities in PDE solutions for efficient numerical methods.
method Data-driven self-supervised learning framework with filtering tasks.
result Proposes filtering methods for raw unlabeled data to improve singularity detection.
Stochastic encoders outperform deterministic ones in 'perfect perceptual quality'.
problem Understanding when stochastic encoders outperform deterministic ones.
method Provided a toy example to illustrate performance.
result Stochastic encoders can significantly outperform deterministic ones in 'perfect perceptual quality'.
BiPE blends intra-segment and inter-segment encodings for better length extrapolation.
problem Improving length extrapolation in language models.
method Bilevel Positional Encoding (BiPE) that separates intra-segment and inter-segment encodings.
result BiPE enhances length extrapolation across various text modalities.
CRATE-MAE learns structured representations from unlabeled data.
problem Learning structured representations from unlabeled data.
method Structured Diffusion with White-Box Transformers.
result CRATE-MAE achieves highly promising performance on large-scale imagery datasets.
Enhanced Bayesian target encoding uses sampling techniques to improve model performance.
problem Improving target encoding for better model performance in machine learning.
method Using sampling techniques in Bayesian target encoding to extract intra-category distribution information.
result Improves generalization and reduces target leakage in machine learning models.
Paper proposes LCP for structural encodings, outperforming existing methods.
problem Improving Graph Neural Networks performance through effective structural encodings.
method Geometric perspective, Local Curvature Profiles (LCP) for structural encodings, combining with global positional encodings, comparing with rewiring techniques.
result LCP significantly outperforms existing structural encodings and combining LCP with global positional encodings improves performance.
New definition reveals encoding explanations that retain predictive power.
problem Challenges in evaluating and identifying encoding explanations.
method Developed a definition of encoding based on conditional dependence.
result Existing evaluation scores do not rank non-encoding explanations correctly, but STRIPE-X does.
Study on encoding neural architectures for NAS, showing impact on performance.
problem Impact of different encodings on NAS performance.
method Formal definition and characterization of encodings, empirical study of encodings' effectiveness.
result Different encodings can significantly affect NAS performance.
Improved OOD detection across various shifts using multi-encoder fusion of RDMs.
problem Out-of-distribution detection across multiple types of distribution shifts.
method Statistical identification of encoder sensitivity, EncMin2L fusion, and Tippett minimum combination.
result Achieves AUROC ≥ 0.94 across four shift types, outperforming state-of-the-art detectors.
Single auto-encoder learns cross-domain image translation.
problem Cross-domain image-to-image translation using a single encoder-decoder architecture.
method Single auto-encoder with independent domain and content encodings.
result Cross-domain mapping achieved without separate encoders.
Binary encoding enables neural networks to extrapolate periodic functions.
problem Extrapolating periodic functions without prior knowledge of their form.
method Normalized Base-2 Encoding (NB2E) for continuous numerical values.
result MLPs using NB2E can successfully extrapolate diverse periodic signals.
This research compares two encoding methods for categorical attributes in machine learning, affecting model fairness.
problem The impact of encoding protected categorical attributes on fairness in machine learning models.
method Comparison of one-hot encoding and target encoding methods.
result Target encoding can lead to more unfair models compared to one-hot encoding due to induced bias.
Quasi-orthonormal encoding reduces high dimensionality for categorical data.
problem High dimensionality and low sample size issues in categorical data encoding.
method Quasi-orthonormal encoding (QOE) for categorical data.
result QOE reduces dimensionality and improves machine learning performance.
For statistical learning, categorical variables in a table are usually considered as discrete entities and encoded separately to feature vectors, e.g., with one-hot encoding. "Dirty" non-curated data gives rise to categorical variables with a very high cardinality but redundancy: several categories reflect the same ent…
In this paper, we propose a refined multi-stage multi-task training strategy to improve the performance of online attention-based encoder-decoder (AED) models. A three-stage training based on three levels of architectural granularity namely, character encoder, byte pair encoding (BPE) based encoder, and attention decod…
The paper identifies saddlepoints in unsupervised auto-encoding neural nets.
problem The risk landscape of unsupervised least squares in auto-encoding neural nets.
method Established an equivalence between unsupervised least squares and principal manifolds, discussed regularization strategies for auto-encoders.
result All non-trivial critical points in auto-encoding are saddlepoints, which are degenerate in overcomplete auto-encoding.
Regularized target encoding beats traditional methods for high cardinality features in ML.
problem Efficiently encoding high cardinality categorical variables for ML algorithms.
method Regularized target encoding compared to traditional encodings like integer and one-hot encoding.
result Regularized target encoding consistently provided the best results in a large-scale benchmark experiment.
We study the role of latent space dimensionality in Wasserstein auto-encoders (WAEs). Through experimentation on synthetic and real datasets, we argue that random encoders should be preferred over deterministic encoders. We highlight the potential of WAEs for representation learning with promising results on a benchmar…
We present a framework to learn privacy-preserving encodings of images that inhibit inference of chosen private attributes, while allowing recovery of other desirable information. Rather than simply inhibiting a given fixed pre-trained estimator, our goal is that an estimator be unable to learn to accurately predict th…
Study efficient neural operator learning using variation spaces.
problem Operator learning using encoder-decoder neural networks.
method Introduce variation space for nonlinear operators, establish approximation bounds.
result Algebraic approximation and learning rates for polynomially decaying input and output encoding errors.
We propose the Wasserstein Auto-Encoder (WAE)---a new algorithm for building a generative model of the data distribution. WAE minimizes a penalized form of the Wasserstein distance between the model distribution and the target distribution, which leads to a different regularizer than the one used by the Variational Aut…
Encoding a sequence of observations is an essential task with many applications. The encoding can become highly efficient when the observations are generated by a dynamical system. A dynamical system imposes regularities on the observations that can be leveraged to achieve a more efficient code. We propose a method to …
Object detection in point clouds is an important aspect of many robotics applications such as autonomous driving. In this paper we consider the problem of encoding a point cloud into a format appropriate for a downstream detection pipeline. Recent literature suggests two types of encoders; fixed encoders tend to be fas…
New method enforces encoder sparsity in HPF for more interpretable feature selection.
problem Lack of encoder sparsity in HPF leads to lack of column-clustering property.
method Enforces encoder sparsity using a generalized additive model (GAM).
result Gains ability to perform feature selection and relates each representation to original features.
DGA and DVGA learn disentangled graph representations to improve graph analysis.
problem Holistic graph auto-encoders fail to capture latent factors effectively.
method Design disentangled graph convolutional network and component-wise flow, impose independence constraints.
result Improved disentangled graph representations enhance graph analysis tasks.
This study predicts parking availability using multi-source data and a self-supervised learning enhanced transformer.
problem Accurate parking availability prediction to support urban planning and management.
method Proposes SST-iTransformer, a self-supervised learning enhanced spatio-temporal inverted transformer, integrating multi-source data.
result SST-iTransformer achieves state-of-the-art performance in parking availability prediction.
STRING improves 2D and 3D position encodings for better performance.
problem Efficient and accurate position encoding for 2D and 3D applications.
method STRING extends Rotary Position Encodings with a unifying theoretical framework, maintaining translation invariance and low computational cost.
result STRING shows substantial gains in open-vocabulary object detection and robotics.
Classical supervised classification tasks search for a nonlinear mapping that maps each encoded feature directly to a probability mass over the labels. Such a learning framework typically lacks the intuition that encoded features from the same class tend to be similar and thus has little interpretability for the learne…
In general, a self-attention mechanism has been applied for speaker embedding encoding. Previous studies focused on training the self-attention in a high-level layer, such as the last pooling layer. However, the effect of low-level features was reduced in the speaker embedding encoding. Therefore, we propose masked cro…
Given a convolutional dictionary underlying a set of observed signals, can a carefully designed auto-encoder recover the dictionary in the presence of noise? We introduce an auto-encoder architecture, termed constrained recurrent sparse auto-encoder (CRsAE), that answers this question in the affirmative. Given an input…
Auto-encoders have emerged as a successful framework for unsupervised learning. However, conventional auto-encoders are incapable of utilizing explicit relations in structured data. To take advantage of relations in graph-structured data, several graph auto-encoders have recently been proposed, but they neglect to reco…