The representations learned by deep neural networks are difficult to interpret in part due to their large parameter space and the complexities introduced by their multi-layer structure. We introduce a method for computing persistent homology over the graphical activation structure of neural networks, which provides acc…
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.
168,878 papers · 148 categories
Trend · papers per month
48 results for “distributed activity representations”
Improved conversion prediction for B2B products using collective online activity trails.
problem Limited information from single user activity trails for B2B ad targeting.
method Introduced relevant users concept and used distributed activity representations to build seed lists.
result Improved conversion prediction AUC by 8.8% using collective activity trails.
We investigate sparse representations for control in reinforcement learning. While these representations are widely used in computer vision, their prevalence in reinforcement learning is limited to sparse coding where extracting representations for new data can be computationally intensive. Here, we begin by demonstrat…
A new criterion for deep active learning selects minimal labeled data points.
problem Efficiently select minimal labeled data points for deep neural networks.
method Diffuses label information over a graph of data representations to switch between exploration and refinement.
result The diffusion-based criterion outperforms existing methods in deep active learning.
New findings clarify the link between distributional closeness and representational similarity.
problem When and why do different neural network representations become similar?
method Identifiability theory, focusing on model families including autoregressive language models.
result Small Kullback-Leibler divergence does not guarantee similar representations.
MAGDiff detects data shifts in neural networks without retraining.
problem Neural networks' sensitivity to data distribution shifts.
method Extracts MAGDiff representations from neural networks to detect shifts.
result MAGDiff representations improve data set shift detection.
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.
ASGN uses active semi-supervised learning to predict molecular properties efficiently.
problem Predicting molecular properties with scarce labeled data and high computational cost.
method ASGN combines a teacher-student framework with active learning to handle joint representation and property learning.
result ASGN achieves remarkable performance in property prediction on public datasets.
Active learning recovers choice model from noisy data.
problem Identifying non-parametric choice models from noisy data.
method Directed acyclic graph (DAG) representation and inclusion-exclusion approach.
result Algorithm more accurately recovers frequent preferences.
Extracting automatically the complex set of features composing real high-dimensional data is crucial for achieving high performance in machine--learning tasks. Restricted Boltzmann Machines (RBM) are empirically known to be efficient for this purpose, and to be able to generate distributed and graded representations of…
An important part of the information gathering and data analysis is to find out what people think about, either a product or an entity. Twitter is an opinion rich social networking site. The posts or tweets from this data can be used for mining people's opinions. The recent surge of activity in this area can be attribu…
This study examines how neural network latent representations correlate with model uncertainty.
problem Detecting model uncertainty in neural networks.
method Empirical verification and analysis of latent representations' distribution and conditional output.
result Deep layers in neural networks can infer uncertainty similar to more computationally expensive methods.
A mathematical theory for understanding when abstract representations emerge in neural networksq-bio.NC
Math theory explains how neural networks learn abstract representations.
problem Understanding how neural networks learn abstract representations.
method Mathematical theory reformulating network optimization into mean field optimization over neural preactivations.
result Abstract representations of latent variables are guaranteed to appear in neural networks trained on tasks that depend on these variables.
Neural networks learn distance-based representations, not just intensity.
problem Understanding how neural networks interpret and learn from internal activations.
method Manipulated ReLU and Absolute Value activations to observe sensitivity to distance and intensity perturbations.
result Neural networks are highly sensitive to small distance-based perturbations, challenging the intensity-based interpretation.
New method embeds bipartite graphs into vectors, overcoming nonlinear challenges.
problem Learning vector representations for bipartite graphs with nonparametric components.
method Semiparametric exponential family distribution, pseudo-likelihood objective, gradient descent.
result Gradient descent achieves linear convergence rate and robust to model misspecification.
Neural network accuracy improves with denser training samples.
problem Improving neural network accuracy on unseen test samples.
method Bounding empirical training error smoothed across activation regions and using it to discard high-risk test samples.
result Discarding high-risk test samples based on error bounds improves prediction accuracy by up to 20%.
New feature representation improves HAR model accuracy.
problem Improving human activity recognition model performance.
method Proposed new feature representation and comparison with existing methods.
result Techniques based on the proposed representation outperform baselines.
BERT-based word embeddings improve active learning for text datasets.
problem Efficiently labelling large text datasets for machine learning.
method Evaluation of text representation mechanisms (BERT vs. bag of words) in active learning.
result BERT-based word embeddings significantly improve active learning performance.
Recently proposed budding tree is a decision tree algorithm in which every node is part internal node and part leaf. This allows representing every decision tree in a continuous parameter space, and therefore a budding tree can be jointly trained with backpropagation, like a neural network. Even though this continuity …
Aims to learn from multiple unpredictable teachers with minimal interaction.
problem Learning from multiple non-deterministic teachers with low interaction cost.
method Develops a framework and an active learning algorithm to estimate a distribution over policy space.
result Significantly reduces interaction with teachers without compromising performance.
SANs use sparse activation functions to compress data representations.
problem Learning meaningful features without considering compression.
method Introduce φ metric, define activation functions, and present SANs.
result SANs achieve small description length and interpretable kernels.
DORA analyzes deep neural networks' internal representations to detect spurious correlations.
problem Detecting spurious correlations in deep neural networks' internal representations.
method DORA uses Extreme-Activation (EA) distance measure to assess representation similarities.
result Identifies internal representations capable of detecting spurious correlations.
Regularized training of an autoencoder typically results in hidden unit biases that take on large negative values. We show that negative biases are a natural result of using a hidden layer whose responsibility is to both represent the input data and act as a selection mechanism that ensures sparsity of the representati…
AL2 progressively penalizes network activations to prevent overfitting.
problem Avoiding overfitting in neural networks with limited training data.
method Progressive Activation Loss (PAL) method to regularize network representations.
result AL2 outperforms traditional regularization methods on benchmark datasets.
End-to-end framework learns precise disparity for activity recognition.
problem Precise portrayal of intraclass disparity in activity recognition.
method Knowledge-directed adversarial learning framework with two competitive encoding distributions.
result Robust and generalizable performance on HAR benchmark datasets.
Emergence of Globally Attracting Fixed Points in Deep Neural Networks With Nonlinear Activationscs.LG
The paper studies how neural networks evolve representations, finding a unique fixed point for nonlinear activations.
problem Understanding how neural networks transform input data across layers.
method Theoretical framework for the evolution of the kernel sequence, using mean-field regime and Hermite polynomials.
result For nonlinear activations, the kernel sequence converges globally to a unique fixed point.
Proposes DWMD for better matching of hidden representations across domains.
problem Measuring data distribution discrepancy between semantically related domains for feature representation matching.
method DWMD, a moment-based probability distribution metric that explicitly orders and weights higher-order moments.
result DWMD is error-free and can strictly reflect distribution differences without feature distribution assumptions.
Model predicts active and passive cosponsorship motivations in Congress.
problem Identifying motivations behind cosponsorship in U.S. Congress.
method Encoder+RGCN model learning from bill texts and speeches.
result F1-score of 0.88 for predicting active and passive cosponsorship.
Sufficient physical activity and restful sleep play a major role in the prevention and cure of many chronic conditions. Being able to proactively screen and monitor such chronic conditions would be a big step forward for overall health. The rapid increase in the popularity of wearable devices provides a significant new…
Spectral measurements reveal hidden representation geometry in language model training.
problem Hidden internal representation in language model training is hard to examine.
method Empirical protocol using activation covariance and per-sample gradient SVD spectra.
result Batch size affects representation geometry, and activation spectra predict token efficiency.
Mean representations of VAEs are correlated but still useful for tasks.
problem Correlation between mean and sampled representations of VAEs.
method Selective posterior collapse to identify active and passive variables.
result Passive variables in mean representations are correlated but uncorrelated in sampled ones.
A novel approach for unsupervised domain adaptation for neural networks is proposed. It relies on metric-based regularization of the learning process. The metric-based regularization aims at domain-invariant latent feature representations by means of maximizing the similarity between domain-specific activation distribu…
VaB-AL: Incorporating Class Imbalance and Difficulty with Variational Bayes for Active Learningcs.LG
Improves Active Learning by considering class imbalance and difficulty.
problem Active Learning's focus on individual samples ignores class distribution and difficulty.
method Proposes a method based on Bayes' rule to incorporate class imbalance, using a Variational Auto Encoder (VAE).
result Significantly outperforms state-of-the-art methods on datasets with heavy data imbalance.
Paper simplifies ANN structure into a functional form.
problem Current ANN structure is complex and difficult to analyze.
method Uses activation integral concept to represent ANN structure as a function.
result Simplified mathematical representation of ANN structure.
SANs use sparse activation functions to minimize model complexity.
problem Model complexity in unsupervised learning.
method Introduce φ metric, define activation functions, present Sparsely Activated Networks (SANs).
result SANs with selected activation functions have small description length and interpretable kernels.
The recently developed variational autoencoders (VAEs) have proved to be an effective confluence of the rich representational power of neural networks with Bayesian methods. However, most work on VAEs use a rather simple prior over the latent variables such as standard normal distribution, thereby restricting its appli…
Unified query framework for active metric learning and classification.
problem Combining representation learning and task-specific goals in machine learning.
method Adaptive selection of nearest neighbor queries using information theoretic criterion.
result Actively selected nearest neighbor queries outperform recent methods in deep metric learning and classification.
Paper proposes HTAF for stable training of binary neural networks.
problem Challenges in training binary neural networks with gradient-based optimization.
method HTAF is a smooth approximation to the Heaviside function that enables stable training.
result HTAF enables stable training of various binary neural networks with gradient-based optimization.
This work tackles representation learning by introducing stochastic competition-based activations.
problem Learning diversified representations in deep learning models.
method Combining information-theoretic arguments with stochastic competition-based activations, using Stochastic Local Winner-Takes-All (LWTA) units.
result The proposed method yields significant discriminative representation learning abilities and allows for a principled investigation of intermediate network representations.
Regularizing activations in neural networks via distribution matching with the Wasserstein metriccs.LG
Proposes PER loss to regularize neural network activations to normal distribution.
problem Improving neural network generalization and training speed.
method Regularizes activations to standard normal distribution via projected error function and Wasserstein distance.
result Minimizes Wasserstein distance between activation distribution and standard normal.
Derives integral formula for ReLU networks with limited weights.
problem Finding optimal neural network weights with limited -norm.
method Derives integral representation formula for shallow ReLU networks under -norm constraint.
result Explicitly solves the least -norm neural network representation for a given function.
Enhances image captioning with novel context combination methods.
problem Improving machine learning for image captioning with structured learning and meaningful interpretation.
method Combines Feature Distribution Composition (FDC), Multiple Role Representation Crossover (MRRC) attention layers, and language decoder.
result Significantly improved image captioning performance (35.3%) and established new standards.
RAAL optimizes black box function optimization with multifidelity models.
problem Time-consuming and unfeasible black box optimization.
method Resource Aware Multifidelity Active Learning (RAAL) for efficient optimization.
result RAAL optimizes black box function optimization with multifidelity models.
Predicting bioactivity and physical properties of small molecules is a central challenge in drug discovery. Deep learning is becoming the method of choice but studies to date focus on mean accuracy as the main metric. However, to replace costly and mission-critical experiments by models, a high mean accuracy is not eno…
With the rapid development of social media sharing, people often need to manage the growing volume of multimedia data such as large scale video classification and annotation, especially to organize those videos containing human activities. Recently, manifold regularized semi-supervised learning (SSL), which explores th…
New test assesses shared brain activity across different cognitive modalities.
problem Determining if different cognitive modalities use overlapping neural representations.
method Formulated a statistical hypothesis testing approach using permutation testing.
result New test (CMPT) has greater statistical power than cross-modal decoding while maintaining low Type I errors.
New dataset and analysis improve evaluation of visual representation models.
problem Insufficient evaluation methods for visual representation models.
method Analyzed five representations and developed a new dataset.
result Models with poor linear classification performance can still perform well on complex tasks.
Unsupervised neural models predict brain activity better than supervised methods.
problem Understanding how the brain represents visual information without direct supervision.
method Built upon PredNet, used RSA to compare PredNet representations to fMRI and MEG data.
result Unsupervised models trained to predict video frames outperform supervised image classification models in predicting brain activity.