Disentangling Dense Embeddings with Sparse Autoencoders
The Australian National University（澳大利亚国立大学）；Stanford University（斯坦福大学）；Columbia University（哥伦比亚大学）；Space Telescope Science Institute（空间望远镜科学研究所）；Johns Hopkins University（约翰斯·霍普金斯大学）
Charles O'Neill, Christine Ye, Kartheik Iyer, John F. Wu
Sparse autoencoders (SAEs) have shown promise in extracting interpretable features from complex neural networks. We present one of the first applications of SAEs to dense text embeddings from large language models, demonstrating their effectiveness in disentangling semantic concepts. By training SAEs on embeddings of over 420,000 scientific paper abstracts from computer science and astronomy, we show that the resulting sparse representations maintain semantic fidelity while offering interpretability. We analyse these learned features, exploring their behaviour across different model capacities and introducing a novel method for identifying ``feature families'' that represent related concepts at varying levels of abstraction. To demonstrate the practical utility of our approach, we show how these interpretable features can be used to precisely steer semantic search, allowing for fine-grained control over query semantics. This work bridges the gap between the semantic richness of dense embeddings and the interpretability of sparse representations. We open source our embeddings, trained sparse autoencoders, and interpreted features, as well as a web app for exploring them.
Rutherford Appleton Laboratory（卢瑟福·阿普尔顿实验室）；Science and Technology Facilities Council（科学与技术设施委员会）
Noting the importance of factorizing (or disentangling) the latent space, we propose a novel, non-probabilistic disentangling framework for autoencoders, based on the principles of symmetry transformations in group-theory. To the best of our knowledge, this is the first deterministic model that is aiming to achieve disentanglement based on autoencoders without regularizers. The proposed model is compared to seven state-of-the-art generative models based on autoencoders and evaluated based on five supervised disentanglement metrics. The experimental results show that the proposed model can have…
University of Oxford（牛津大学）；Alan Turing Institute（艾伦·图灵研究所）
In clustering we normally output one cluster variable for each datapoint. However it is not necessarily the case that there is only one way to partition a given dataset into cluster components. For example, one could cluster objects by their colour, or by their type. Different attributes form a hierarchy, and we could wish to cluster in any of them. By disentangling the learnt latent representations of some dataset into different layers for different attributes we can then cluster in those latent spaces. We call this "disentangled clustering". Extending Variational Ladder Autoencoders (Zhao et…
In the last few years there have been important advancements in generative models with the two dominant approaches being Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs). However, standard Autoencoders (AEs) and closely related structures have remained popular because they are easy to train and adapt to different tasks. An interesting question is if we can achieve state-of-the-art performance with AEs while retaining their good properties. We propose an answer to this question by introducing a new model called Y-Autoencoder (Y-AE). The structure and training procedure…
The ability to extract generative parameters from high-dimensional fields of data in an unsupervised manner is a highly desirable yet unrealized goal in computational physics. This work explores the use of variational autoencoders (VAEs) for non-linear dimension reduction with the specific aim of {\em disentangling} the low-dimensional latent variables to identify independent physical parameters that generated the data. A disentangled decomposition is interpretable, and can be transferred to a variety of tasks including generative modeling, design optimization, and probabilistic reduced order …
The use of well-disentangled representations offers many advantages for downstream tasks, e.g. an increased sample efficiency, or better interpretability. However, the quality of disentangled interpretations is often highly dependent on the choice of dataset-specific hyperparameters, in particular the regularization strength. To address this issue, we introduce DAVA, a novel training procedure for variational auto-encoders. DAVA completely alleviates the problem of hyperparameter selection. We compare DAVA to models with optimal hyperparameters. Without any hyperparameter tuning, DAVA is compe…
We develop a generalisation of disentanglement in VAEs---decomposition of the latent representation---characterising it as the fulfilment of two factors: a) the latent encodings of the data having an appropriate level of overlap, and b) the aggregate encoding of the data conforming to a desired structure, represented through the prior. Decomposition permits disentanglement, i.e. explicit independence between latents, as a special case, but also allows for a much richer class of properties to be imposed on the learnt representation, such as sparsity, clustering, independent subspaces, or even i…
School of Computer Science, University of Waterloo（滑铁卢大学计算机科学学院）；Systems Design Engineering, University of Waterloo（滑铁卢大学系统设计工程系）；Department of Statistics and Actuarial Science, University of Waterloo（滑铁卢大学统计与精算科学系）
The problem of feature disentanglement has been explored in the literature, for the purpose of image and video processing and text analysis. State-of-the-art methods for disentangling feature representations rely on the presence of many labeled samples. In this work, we present a novel method for disentangling factors of variation in data-scarce regimes. Specifically, we explore the application of feature disentangling for the problem of supervised classification in a setting where few labeled samples exist, and there are no unlabeled samples for use in unsupervised training. Instead, a simila…
Recently, it has been observed that when representations are learnt in a way that encourages sparsity, improved performance is obtained on classification tasks. These methods involve combinations of activation functions, sampling steps and different kinds of penalties. To investigate the effectiveness of sparsity by itself, we propose the k-sparse autoencoder, which is an autoencoder with linear activation function, where in hidden layers only the k highest activities are kept. When applied to the MNIST and NORB datasets, we find that this method achieves better classification results than den…
College of Economics and Management, Shandong University of Science and Technology（山东科技大学经济管理学院）
We propose a novel filter for sparse big data, called an integrated autoencoder (IAE), which utilizes auxiliary information to mitigate data sparsity. The proposed model achieves an appropriate balance between prediction accuracy, convergence speed, and complexity. We implement experiments on a GPS trajectory dataset, and the results demonstrate that the IAE is more accurate and robust than some state-of-the-art methods.
A variational autoencoder (VAE) is a probabilistic machine learning framework for posterior inference that projects an input set of high-dimensional data to a lower-dimensional, latent space. The latent space learned with a VAE offers exciting opportunities to develop new data-driven design processes in creative disciplines, in particular, to automate the generation of multiple novel designs that are aesthetically reminiscent of the input data but that were unseen during training. However, the learned latent space is typically disorganized and entangled: traversing the latent space along a sin…
Disentangling complex data to its latent factors of variation is a fundamental task in representation learning. Existing work on sequential disentanglement mostly provides two factor representations, i.e., it separates the data to time-varying and time-invariant factors. In contrast, we consider multifactor disentanglement in which multiple (more than two) semantic disentangled components are generated. Key to our approach is a strong inductive bias where we assume that the underlying dynamics can be represented linearly in the latent space. Under this assumption, it becomes natural to exploit…
Recent approaches in the theoretical analysis of model-based deep learning architectures have studied the convergence of gradient descent in shallow ReLU networks that arise from generative models whose hidden layers are sparse. Motivated by the success of architectures that impose structured forms of sparsity, we introduce and study a group-sparse autoencoder that accounts for a variety of generative models, and utilizes a group-sparse ReLU activation function to force the non-zero units at a given layer to occur in blocks. For clustering models, inputs that result in the same group of active…
Matrix factorization (MF) and Autoencoder (AE) are among the most successful approaches of unsupervised learning. While MF based models have been extensively exploited in the graph modeling and link prediction literature, the AE family has not gained much attention. In this paper we investigate both MF and AE's application to the link prediction problem in sparse graphs. We show the connection between AE and MF from the perspective of multiview learning, and further propose MF+AE: a model training MF and AE jointly with shared parameters. We apply dropout to training both the MF and AE parts, …
SAE-FD: Sparse Autoencoder Feature Distillation for Continual Learning of Large Language Models
The Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））；Nanjing University of Aeronautics and Astronautics（南京航空航天大学）；The 63rd Research Institute, National University of Defense Technology, Nanjing（国防科技大学第六十三研究所，南京）
Evaluating Sparse Autoencoders: From Shallow Design to Matching Pursuit
Harvard University（哈佛大学）；Kempner Institute（凯姆纳研究所）；Department of Electrical and Computer Engineering, University of Alberta（阿尔伯塔大学电气与计算机工程系）；CBS-NTT Program in Physics of Intelligence, Harvard University（哈佛大学物理智能项目）；Alberta Machine Intelligence Institute (Amii)（阿尔伯塔机器智能研究所）；Neuroscience and Mental Health Institute, University of Alberta（阿尔伯塔大学神经科学与心理健康研究所）
Can sparse autoencoders make sense of gene expression latent variable models?
Eric and Wendy Schmidt Center, Broad Institute of MIT and Harvard（埃里克和文迪斯中心，哈佛-麻省理工Broad研究所）；Department of Computer Science, University of Copenhagen（哥本哈根大学计算机科学系）
Several recent results in machine learning have established formal connections between autoencoders---artificial neural network models that attempt to reproduce their inputs---and other coding models like sparse coding and K-means. This paper explores in depth an autoencoder model that is constructed using rectified linear activations on its hidden units. Our analysis builds on recent results to further unify the world of sparse linear coding models. We provide an intuitive interpretation of the behavior of these coding models and demonstrate this intuition using small, artificial datasets wit…
Beyond Activation Patterns: A Weight-Based Out-of-Context Explanation of Sparse Autoencoder Features
State Key Laboratory of General Artificial Intelligence, School of Intelligence Science and Technology, Peking University（通用人工智能国家重点实验室，智能科学与技术学院，北京大学）
Denoising, detrending, deconvolution: usual restoration tasks, traditionally decoupled. Coupled formulations entail complex ill-posed inverse problems. We propose PENDANTSS for joint trend removal and blind deconvolution of sparse peak-like signals. It blends a parsimonious prior with the hypothesis that smooth trend and noise can somewhat be separated by low-pass filtering. We combine the generalized quasi-norm ratio SOOT/SPOQ sparse penalties $\ell_p/\ell_q$ with the BEADS ternary assisted source separation algorithm. This results in a both convergent and efficient tool, with a novel Trust-R…
School of Microelectornics and Communication Engineering, Chongqing University（重庆大学微电子与通信工程学院）
Deep learning is a kind of feature learning method with strong nonliear feature transformation and becomes more and more important in many fields of artificial intelligence. Deep autoencoder is one representative method of the deep learning methods, and can effectively extract abstract the information of datasets. However, it does not consider the complementarity between the deep features and original features during deep feature transformation. Besides, it suffers from small sample problem. In order to solve these problems, a novel deep autoencoder - hybrid feature embedded stacked sparse aut…
EURECOM（欧洲通信研究院）；University of California, Irvine（加州大学欧文分校）
Autoencoders and their variants are among the most widely used models in representation learning and generative modeling. However, autoencoder-based models usually assume that the learned representations are i.i.d. and fail to capture the correlations between the data samples. To address this issue, we propose a novel Sparse Gaussian Process Bayesian Autoencoder (SGPBAE) model in which we impose fully Bayesian sparse Gaussian Process priors on the latent space of a Bayesian Autoencoder. We perform posterior estimation for this model via stochastic gradient Hamiltonian Monte Carlo. We evaluate …
Redwood Center for Theoretical Neuroscience（雷德伍德理论神经科学中心）；Berkeley AI Research（伯克利人工智能研究院）；University of California, Berkeley（加州大学伯克利分校）；Facebook AI Research（脸书人工智能研究院）
Discrete spatial patterns and their continuous transformations are two important regularities contained in natural signals. Lie groups and representation theory are mathematical tools that have been used in previous works to model continuous image transformations. On the other hand, sparse coding is an important tool for learning dictionaries of patterns in natural signals. In this paper, we combine these ideas in a Bayesian generative model that learns to disentangle spatial patterns and their continuous transformations in a completely unsupervised manner. Images are modeled as a sparse super…
University of Gothenburg（哥德堡大学）；Chalmers University of Technology（查尔姆斯理工大学）
High-dimensional data sets are often analyzed and explored via the construction of a latent low-dimensional space which enables convenient visualization and efficient predictive modeling or clustering. For complex data structures, linear dimensionality reduction techniques like PCA may not be sufficiently flexible to enable low-dimensional representation. Non-linear dimension reduction techniques, like kernel PCA and autoencoders, suffer from loss of interpretability since each latent variable is dependent of all input dimensions. To address this limitation, we here present path lasso penalize…
LIPN（LIPN实验室）；Université Sorbonne Paris Nord（索邦巴黎北大学）；INRIA Paris（法国国家信息与自动化研究所巴黎分部）
We present an unsupervised method to obtain disentangled representations of sentences that single out semantic content. Using modified Transformers as building blocks, we train a Variational Autoencoder to translate the sentence to a fixed number of hierarchically structured latent variables. We study the influence of each latent variable in generation on the dependency structure of sentences, and on the predicate structure it yields when passed through an Open Information Extraction model. Our model could separate verbs, subjects, direct objects, and prepositional objects into latent variable…
The classical sparse coding model represents visual stimuli as a linear combination of a handful of learned basis functions that are Gabor-like when trained on natural image data. However, the Gabor-like filters learned by classical sparse coding far overpredict well-tuned simple cell receptive field (SCRF) profiles. A number of subsequent models have either discarded the sparse dictionary learning framework entirely or have yet to take advantage of the surge in unrolled, neural dictionary learning architectures. A key missing theme of these updates is a stronger notion of \emph{structured spa…
University of Louisville（路易斯维尔大学）；University of Social Science（社会科学大学）；Information Technology Institute（信息技术研究所）
We demonstrate a new deep learning autoencoder network, trained by a nonnegativity constraint algorithm (NCAE), that learns features which show part-based representation of data. The learning algorithm is based on constraining negative weights. The performance of the algorithm is assessed based on decomposing data into parts and its prediction performance is tested on three standard image data sets and one text dataset. The results indicate that the nonnegativity constraint forces the autoencoder to learn features that amount to a part-based representation of data, while improving sparsity and…
This paper aims to improve the feature learning in Convolutional Networks (Convnet) by capturing the structure of objects. A new sparsity function is imposed on the extracted featuremap to capture the structure and shape of the learned object, extracting interpretable features to improve the prediction performance. The proposed algorithm is based on organizing the activation within and across featuremap by constraining the node activities through $\ell_{2}$ and $\ell_{1}$ normalization in a structured form.
Philips Medical Systems Technologies Ltd.（飞利浦医疗系统科技有限公司）
Deep sparse auto-encoders with mixed structure regularization (MSR) in addition to explicit sparsity regularization term and stochastic corruption of the input data with Gaussian noise have the potential to improve unsupervised abnormality detection. Unsupervised abnormality detection based on identifying outliers using deep sparse auto-encoders is a very appealing approach for medical computer aided detection systems as it requires only healthy data for training rather than expert annotated abnormality. In the task of detecting coronary artery disease from Coronary Computed Tomography Angiogr…
Ostwestfalen-Lippe University of Applied Sciences（东威斯特伐利亚-利珀应用科学大学）；Phoenix Contact Electronics GmbH（菲尼克斯电气电子有限公司）
Digitalization leads to data transparency for production systems that we can benefit from with data-driven analysis methods like neural networks. For example, automated anomaly detection enables saving resources and optimizing the production. We study using rarely occurring information about labeled anomalies into Variational Autoencoder neural network structures to overcome information deficits of supervised and unsupervised approaches. This method outperforms all other models in terms of accuracy, precision, and recall. We evaluate the following methods: Principal Component Analysis, Isolati…
In this paper, we propose a winner-take-all method for learning hierarchical sparse representations in an unsupervised fashion. We first introduce fully-connected winner-take-all autoencoders which use mini-batch statistics to directly enforce a lifetime sparsity in the activations of the hidden units. We then propose the convolutional winner-take-all autoencoder which combines the benefits of convolutional architectures and autoencoders for learning shift-invariant sparse representations. We describe a way to train convolutional autoencoders layer by layer, where in addition to lifetime spars…
