A new method for the unsupervised learning of sparse representations using autoencoders is proposed and implemented by ordering the output of the hidden units by their activation value and progressively reconstructing the input in this order. This can be done efficiently in parallel with the use of cumulative sums and …
On-device research index
arXiv research
A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
168,695 papers · 148 categories
Trend · papers per month
48 results for “activation reconstruction”
IMPACT optimizes LLM compression by focusing on activation importance, reducing model size up to 55.4%.
problem Resource constraints in deploying large language models (LLMs).
method IMPACT integrates activation importance into low-rank compression, optimizing for both size and accuracy.
result IMPACT achieves up to 55.4% greater model size reduction while maintaining comparable or better accuracy.
BatchTopK SAEs improve GPT-2 and Gemma activations with adjustable sparsity.
problem Interpreting language model activations using sparse autoencoders.
method Adapting the TopK constraint to the batch-level, allowing variable number of active latents per sample.
result BatchTopK SAEs consistently outperform TopK SAEs in reconstructing GPT-2 and Gemma activations.
Cross-Layer Subspace Coupling for LLM Compression: A Unifying Framework and Its Empirical Limitscs.LG
Unified SVD compression fails in practical tasks, highlighting the importance of per layer activation reconstruction.
problem The failure of a unified SVD compression method in practical tasks like perplexity and accuracy.
method Unified optimization problem for SVD based compression methods, focusing on cross-layer coupling.
result Downstream metrics like perplexity and accuracy degrade severely compared to standard per layer SVD LLM.
Recent literature on unsupervised learning focused on designing structural priors with the aim of learning meaningful features, but without considering the description length of the representations. In this thesis, first we introduce the metric that evaluates unsupervised models based on their reconstruction …
Previous literature on unsupervised learning focused on designing structural priors with the aim of learning meaningful features. However, this was done without considering the description length of the learned representations which is a direct and unbiased measure of the model complexity. In this paper, first we intro…
New study analyzes security of neural network data reconstruction attacks.
problem Data reconstruction attacks pose a threat to private training data.
method Analyzes security boundary of data reconstruction attacks via neuron exclusivity state.
result Characterizes insecure/secure boundary of data reconstruction attacks.
A fall is an abnormal activity that occurs rarely, so it is hard to collect real data for falls. It is, therefore, difficult to use supervised learning methods to automatically detect falls. Another challenge in using machine learning methods to automatically detect falls is the choice of engineered features. In this p…
This work improves data reconstruction methods by ensuring unique solutions and refining optimization.
problem Ensuring unique solutions and optimizing reconstruction from KKT conditions.
method Discussion of sufficient conditions for unique solutions and introduction of sample splitting for optimization.
result Sample splitting improves reconstruction performance across various methods.
Develops Active Fourier Auditor to estimate ML model properties without reconstructing them.
problem Verifying and auditing properties of Machine Learning models in real-world applications.
method A new framework that quantifies ML model properties using Fourier coefficients, without reconstructing the model.
result Active Fourier Auditor (AFA) is more accurate and sample-efficient than baselines for estimating robustness, individual fairness, and group fairness.
Bayesian framework optimizes 3D view selection for specific tasks.
problem Optimizing 3D view selection for specific tasks in reconstruction.
method Bayesian decision theory, prior and posterior distributions, stochastic surface reconstruction.
result Framework achieves superior performance with fewer views.
Paper reconstructs training data from a single gradient query.
problem Privacy threats in federated learning due to model gradients.
method Provable attack using tensor decomposition.
result Training samples can be fully reconstructed from a single gradient query.
Improving Model-Based Control and Active Exploration with Reconstruction Uncertainty Optimizationcs.LG
Model based predictions of future trajectories of a dynamical system often suffer from inaccuracies, forcing model based control algorithms to re-plan often, thus being computationally expensive, suboptimal and not reliable. In this work, we propose a model agnostic method for estimating the uncertainty of a model?s pr…
New method uncovers small but significant local activities in time-series data.
problem Reconstructing small but important local activities in time-series data.
method Neural state-space models with latent causal-effect disentanglement.
result Demonstrated proof-of-concept on reconstructing ectopic foci in cardiac electrical propagation.
Dynamics and function of neuronal networks are determined by their synaptic connectivity. Current experimental methods to analyze synaptic network structure on the cellular level, however, cover only small fractions of functional neuronal circuits, typically without a simultaneous record of neuronal spiking activity. H…
Survey of deep learning methods for fMRI natural image reconstruction.
problem Reconstructing natural images from fMRI brain activity.
method Survey of deep learning approaches, including architectural design, datasets, and evaluation metrics.
result Performance evaluation across standardized metrics.
Method detects insider trading using trading data and dimensionality reduction.
problem Identifying insider trading in large datasets.
method Unsupervised machine learning, principal component analysis, autoencoders.
result Identifies suspicious trading behavior based on reconstruction errors.
Fast reconstruction of degenerate populations of conductance-based neuron models from spike timesq-bio.NC
Method reconstructs neuron models from spike times efficiently.
problem Reconstructing neuron models from spike times in degenerate populations.
method Combining deep learning with DICs to map spike times to DIC densities and generate degenerate CBM populations.
result Fast and scalable reconstruction of degenerate populations from spike recordings.
This paper presents a novel method for the reconstruction of a neural network connectivity using calcium fluorescence data. We introduce a fast unsupervised method to integrate different networks that reconstructs structural connectivity from neuron activity. Our method improves the state-of-the-art reconstruction meth…
SAEs struggle with curved activation manifolds, revealing layer-dependent scaling laws.
problem Sparse autoencoders' reconstruction error varies across layers, not fitting existing scaling laws.
method Cross-layer study of 844 SAE checkpoints, fitting and regressing on manifold geometry.
result Manifold geometry predicts layer-dependent width exponents in SAEs, with transferable coefficients.
BrainSurfCNN predicts task contrasts from resting-state fingerprints, improving accuracy over baseline.
problem Predicting task-evoked activity from resting-state functional connectivity.
method Surface-based convolutional neural network (BrainSurfCNN) with reconstructive-contrastive loss.
result Significantly improved accuracy in predicting task contrasts over baseline.
Deep adversarial neural decodingq-bio.NC
Here, we present a novel approach to solve the problem of reconstructing perceived stimuli from brain responses by combining probabilistic inference with deep learning. Our approach first inverts the linear transformation from latent features to brain responses with maximum a posteriori estimation and then inverts the …
Binary autoencoder with sparse hidden layer preserves information and zero reconstruction error.
problem Preserving information and zero reconstruction error in binary neural networks.
method Binary autoencoder with random binary weights, sparse hidden layer, and varying neuron thresholds.
result Zero reconstruction error for any input with a large hidden layer and varying neuron thresholds.
This paper proposes an end-to-end approach for single-channel speaker-independent multi-speaker speech separation, where time-frequency (T-F) masking, the short-time Fourier transform (STFT), and its inverse are represented as layers within a deep network. Previous approaches, rather than computing a loss on the recons…
Iterative learning to infer approaches have become popular solvers for inverse problems. However, their memory requirements during training grow linearly with model depth, limiting in practice model expressiveness. In this work, we propose an iterative inverse model with constant memory that relies on invertible networ…
Algorithm reconstructs triangle-free networks from data, certifying correctness.
problem Reconstructing triangle-free dynamic networks from observational data.
method Developed an algorithm for triangle-free networks, providing guarantees on correctness.
result Algorithm either certifies correctness or outputs a sparser graph with no false positives.
Spreading processes are often modelled as a stochastic dynamics occurring on top of a given network with edge weights corresponding to the transmission probabilities. Knowledge of veracious transmission probabilities is essential for prediction, optimization, and control of diffusion dynamics. Unfortunately, in most ca…
Ill-posed inverse problems in imaging remain an active research topic in several decades, with new approaches constantly emerging. Recognizing that the popular dictionary learning and convolutional sparse coding are both essentially modeling the high-frequency component of an image, which convey most of the semantic in…
The superior temporal gyrus (STG) region of cortex critically contributes to speech recognition. In this work, we show that a proposed WaveNet, with limited available data, is able to reconstruct speech stimuli from STG intracranial recordings. We further investigate the impulse response of the fitted model for each re…
In the context of dynamic emission tomography, the conventional processing pipeline consists of independent image reconstruction of single time frames, followed by the application of a suitable kinetic model to time activity curves (TACs) at the voxel or region-of-interest level. The relatively new field of 4D PET dire…
We consider the problem of offline, pool-based active semi-supervised learning on graphs. This problem is important when the labeled data is scarce and expensive whereas unlabeled data is easily available. The data points are represented by the vertices of an undirected graph with the similarity between them captured b…
We investigate active learning by pairwise similarity over the leaves of trees originating from hierarchical clustering procedures. In the realizable setting, we provide a full characterization of the number of queries needed to achieve perfect reconstruction of the tree cut. In the non-realizable setting, we rely on k…
Image reconstruction in low-count PET is particularly challenging because gammas from natural radioactivity in Lu-based crystals cause high random fractions that lower the measurement signal-to-noise-ratio (SNR). In model-based image reconstruction (MBIR), using more iterations of an unregularized method may increase t…
Geometric Approaches to Increase the Expressivity of Deep Neural Networks for MR Reconstructioncs.CV
Enhances deep neural networks for MRI reconstruction by increasing expressivity.
problem Balancing network complexity and performance in deep learning MRI reconstruction.
method Geometric approach using bootstrapping and subnetwork aggregation with attention module.
result Significant improvement in MRI reconstruction performance with minimal complexity increase.
The Rectified Linear Unit (ReLU) is a foundational activation function in artficial neural networks. Recent literature frequently misattributes its origin to the 2018 (initial) version of this paper, which exclusively investigated ReLU at the classification layer. This paper formally corrects the citation record by tra…
Variational methods for revealing visual concepts learned by convolutional neural networks have gained significant attention during the last years. Being based on noisy gradients obtained via back-propagation such methods require the application of regularization strategies. We present a mathematical framework unifying…
Reconstructing network connectivity from the collective dynamics of a system typically requires access to its complete continuous-time evolution although these are often experimentally inaccessible. Here we propose a theory for revealing physical connectivity of networked systems only from the event time series their i…
Kinetic Compressive Sensingphysics.med-ph
Parametric images provide insight into the spatial distribution of physiological parameters, but they are often extremely noisy, due to low SNR of tomographic data. Direct estimation from projections allows accurate noise modeling, improving the results of post-reconstruction fitting. We propose a method, which we name…
Predicts trainability of deep neural networks using reconstruction entropy.
problem Predicting the initial conditions for trainability of deep neural networks.
method Cascade of auxiliary networks to reconstruct input from activation layers, computing relative entropy.
result Predicts trainability of deep feedforward networks on various datasets with a single epoch.
We present a graph-based semi-supervised learning (SSL) method for learning edge flows defined on a graph. Specifically, given flow measurements on a subset of edges, we want to predict the flows on the remaining edges. To this end, we develop a computational framework that imposes certain constraints on the overall fl…
Enhances 2D face recognition with 3D features using active illumination.
problem Improving robustness of 2D face recognition to spoofing attacks and low-light conditions.
method Projecting a high spatial frequency pattern onto the face to recover 3D information and a 2D image simultaneously.
result Significantly boosts face recognition performance and dramatically improves robustness to spoofing attacks.
The paper proposes methods to find a shared active subspace for multivariate vector-valued functions.
problem Minimizing the deviation between function evaluations in the original and reconstructed spaces.
method Manipulating gradients or SPD matrices to identify a shared structure.
result Summing SPD matrices often identifies the best shared active subspace.
This paper develops a novel deep recurrent neural network for sequential signal reconstruction.
problem Sequential signal reconstruction from low-dimensional measurements.
method Unfolding a reweighted - minimization algorithm to design a deep recurrent neural network.
result The proposed reweighted-RNN significantly outperforms existing RNN models in sequential frame reconstruction.
A new algorithm speeds up EEG source localization using regularization.
problem Challenging inverse problem in mapping EEG readings to brain activity.
method Formulated as a graphical generalized elastic net inverse problem, solved with a variable projected algorithm (VPAL).
result VPAL provides faster and more accurate EEG source localization compared to existing methods.
Brain source imaging is an important method for noninvasively characterizing brain activity using Electroencephalogram (EEG) or Magnetoencephalography (MEG) recordings. Traditional EEG/MEG Source Imaging (ESI) methods usually assume that either source activity at different time points is unrelated, or that similar spat…
Reduces data leakage in distributed deep learning models.
problem Prevents reconstruction of sensitive raw data patterns during client communications.
method Reduces distance correlation between raw data and learned representations.
result Resilient to reconstruction attacks while maintaining model accuracy.
A fast method for Lasso and Logistic Lasso problems.
problem Solving Lasso and Logistic Lasso regression problems efficiently.
method Iterative active set approach using solver updates.
result 31.41 times faster on average for compressed sensing.
A Graph Feature Auto-Encoder for the Prediction of Unobserved Node Features on Biological Networksq-bio.QM
Graph auto-encoder predicts unobserved node features from biological networks and omics data.
problem Integrating biological networks and continuous node features for better prediction.
method Graph neural networks and feature auto-encoders trained on feature reconstruction.
result Graph feature auto-encoder outperforms auto-encoders trained on graph reconstruction for predicting unobserved node features.