Research
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.

169,181 papers · 148 categories

Trend · papers per month

52103155206 · Jun 202019922001200920182026
48 results for influence vectors

Unified approach for influence maximization using diffusion cascade representations.

problem Influence maximization on networks with diffusion cascades.
method Multi-task neural network learning influencer and susceptible vectors; greedy algorithm for influence maximization.
result IMINFECTOR outperforms other methods in efficiency and seed set quality.

Establishes statistical and computational bounds for influence diagnostics.

problem Identifying influential datapoints or subsets in machine learning models.
method Finite-sample statistical bounds and computational complexity for influence functions and approximate maximum influence perturbations.
result Established statistical and computational guarantees for influence diagnostics.

Many networks are complex dynamical systems, where both attributes of nodes and topology of the network (link structure) can change with time. We propose a model of co-evolving networks where both node at- tributes and network structure evolve under mutual influence. Specifically, we consider a mixed membership stochas…

2011-06-14abs ↗pdf ↗

How can we explain the predictions of a black-box model? In this paper, we use influence functions -- a classic technique from robust statistics -- to trace a model's prediction through the learning algorithm and back to its training data, thereby identifying training points most responsible for a given prediction. To …

2017-03-14abs ↗pdf ↗

Paper proposes a new model to infer topic-based connection structures from noisy adjacency matrices.

problem Traditional community detection assumes static connection structures, but real-world connections vary based on topic.
method Introduces latent model with influence and receptivity vectors for each node, estimating topic distributions from observed data.
result The model can estimate topic-based connection structures with theoretical guarantees and outperforms existing methods.

Influence functions help study large language model generalization, revealing surprising decay patterns.

problem Understanding and mitigating risks in large language models (LLMs).
method Eigenvalue-corrected Kronecker-Factored Approximation (EK-FAC) to scale influence functions to LLMs.
result Influences decay to near-zero when key phrases order is flipped, revealing a surprising limitation.

We study the problem of learning the support of transition matrix between random processes in a Vector Autoregressive (VAR) model from samples when a subset of the processes are latent. It is well known that ignoring the effect of the latent processes may lead to very different estimates of the influences among observe…

2017-02-27abs ↗pdf ↗

Improves understanding of neural network predictions using influence functions.

problem Challenges in understanding neural network predictions.
method Utilized NTK theory to calculate influence functions for over-parameterized neural networks.
result Proved that the approximation error of IF can be arbitrarily small in the over-parameterized regime.

Simple linear models reveal complex cryptocurrency networks.

problem Understanding complex causal networks in cryptocurrency markets.
method Multivariate linear models to infer financial networks from cryptocurrency price series.
result Simple linear models can create informative cryptocurrency networks reflecting economic intuition.

The paper characterizes Riemannian manifolds using concircular vector fields and a connecting function.

problem Characterizing Riemannian manifolds using concircular vector fields.
method Introducing a connecting function that links concircular vector fields to potential functions.
result The connecting function is crucial for characterizing nn-sphere and Euclidean space.

Study noncommutative deformations of Calabi-Yau threefolds.

problem Understanding the geometry of Calabi-Yau threefolds under noncommutative deformations.
method Analyzing the influence of Poisson structures on quantum moduli spaces.
result The choice of Poisson structure significantly affects the geometry of quantum moduli spaces.

This paper compares different activation functions in GLVQ models.

problem Improving performance of GLVQ models using different activation functions.
method Investigates and compares ReLU, sigmoid, and swish activation functions in GLVQ models.
result Different activation functions have varying impacts on GLVQ model performance.

Study examines how event rate affects bankruptcy prediction model performance.

problem Effect of event rate on bankruptcy prediction model performance.
method Oversampled event rates from 0.12% to 50%, developed and evaluated 7 models.
result Bayesian Network is least sensitive to event rate, SVM most sensitive.

Paper tackles hyper-gradient estimation in decentralized FL over time-varying networks.

problem Excessive communication costs and inability to use robust networks.
method Introduces an optimality condition and uses Push-Sum for averaging model parameters and gradients over time-varying directed networks.
result Derives a hyper-gradient estimator that operates over time-varying directed networks and converges to the true hyper-gradient.

Improved fuzzy support vector machine for stock price trend forecasting.

problem Weak performance of traditional support vector machines in handling fuzzy and noisy data.
method Proposed a novel advanced fuzzy support vector machine (NA-FSVM) to improve precision.
result Improved model precision in predicting stock price trends.

ASTRA improves TDA by more accurately approximating iHVP.

problem Improving insights into training data attribution.
method ASTRA uses EKFAC-preconditioner on Neumann series iterations to accurately approximate iHVP.
result Improving iHVP approximation significantly improves TDA performance.

This paper uses SVM to predict stock market trends from financial news.

problem Predicting stock market trends using text mining and sentiment analysis.
method Text mining, sentiment analysis, support vector machine (SVM), parameter optimization.
result SVM models show significant influence of news on stock market, with parameter G having the main effect.

Heteroencoders improve chemical latent space diversity and molecular generation.

problem Improving chemical latent space properties and diversity in autoencoders.
method Employing SMILES enumeration for encoder or decoder, training RNNs with LSTM, and using QSAR models.
result Heteroencoders yield more diverse latent spaces and better molecular generation.

Study examines how wind affects shortest paths on Finsler manifolds.

problem Investigating shortest paths in Finsler manifolds with wind effects.
method Coordinate-free approach to compare isoparametric functions and mean curvatures.
result Mean curvatures differ in presence and absence of wind on Finsler manifolds.

A generalization of expectiles for d-dimensional multivariate distribution functions is introduced. The resulting geometric expectiles are unique solutions to a convex risk minimization problem and are given by d-dimensional vectors. They are well behaved under common data transformations and the corresponding sample v…

2017-04-05abs ↗pdf ↗

New architecture uses vector fields to move data in neural networks.

problem Improving neural network architectures and performance.
method Exploring vector fields as a new interpretation of neural networks, proposing Vector Fields Neural Networks (VFNN). Using Euler's method to solve ODEs and Gaussian vector fields.
result VFNN shows comparable or better results than basic models for different datasets.

New method reveals true causal functions in nonlinear time series, not just scores.

problem Causal discovery in nonlinear time series often uses scalar edge scores, which hide true function-valued causal influence.
method Formalized function-valued causal influence for additive, contribution-decomposable architectures. Introduced a practical framework based on ICE for estimating causal response functions directly from trained models.
result Edges with indistinguishable scalar scores can exhibit qualitatively different functional behaviors.

Proposes an evolutionary approach to fitting acyclic VAR models.

problem Cycles in multivariate time series systems obscure hierarchical analysis.
method Evolutionary approach to fitting acyclic VAR processes with hierarchical representation.
result Outperforms unconstrained models and captures key structural properties.

Paper develops a non-parametric model to estimate influence networks in high-dimensional time series.

problem Estimating influence networks in high-dimensional time series with many variables.
method Non-parametric sparse additive model (SpAM) using β and φ-mixing properties of Markov chains and empirical process techniques for RKHSs.
result Sharp upper bounds on mean-squared error for estimating influence networks.

Develops a method to audit indirect feature influence in complex models.

problem Auditing indirect feature influence in complex, black-box models.
method Disentangled influence audits using disentangled representations.
result Can detect proxy features and show which ones affect model outcomes most.

Proposes a new model to capture joint influence of correlated events on user search behavior.

problem Real-world events influence each other and pose joint influence on user search behavior, not independent.
method Joint Influence Model based on Multivariate Hawkes Process.
result The model captures the temporal dynamics of joint influence and outperforms baseline methods.

Dynamic Influence Tracker measures changing sample importance during model training.

problem Static influence measurements during training overlook how sample importance varies over time.
method Dynamic Influence Tracker (DIT) captures time-varying sample influence across arbitrary time windows.
result DIT reveals distinct learning phases with shifting priorities and detects corrupted samples more efficiently.

Study compares machine learning models for tourism demand forecasting in Spain.

problem Improving tourism demand forecasting accuracy with machine learning.
method Comparison of Support Vector Regression (SVR) and Neural Network (NN) models with a linear model benchmark.
result SVR with Gaussian radial basis function kernel outperforms other models for long forecast horizons.

RelatIF selects more intuitive training examples for explaining model predictions.

problem Influence functions identify outliers as explanatory examples, leading to poor explanations.
method RelatIF separates global and local influence, optimizing for local relative to global effects.
result Examples selected by RelatIF are more intuitive than those from influence functions.

We propose a mathematical procedure for finding informed trader activities in European-style options and their underlying asset. The regression model (9) with moving average component was written. Being added to it ARMA-process for log-price differences of underlying asset, the generalized model is written as Vector AR…

2014-03-13abs ↗pdf ↗

Study shows how order flow at multiple price levels affects stock prices.

problem Understanding how order flow at different price levels influences stock prices.
method Fit a linear relationship between multi-level order-flow imbalance (MLOFI) and mid-price changes using high-quality data.
result The inclusion of more price levels in MLOFI improves the fit with mid-price changes.

Complete criterion for VoI in multi-decision influence diagrams established.

problem Analyzing safety and fairness properties of AI systems using influence diagrams.
method Introduced ID homomorphisms and Tree of Systems to prove properties of multi-decision influence diagrams.
result First complete graphical criterion for VoI in influence diagrams with multiple decisions.

Influence functions are inaccurate in deep learning models, especially for deeper networks.

problem Inaccuracies in influence functions in deep learning models.
method Empirical study of influence functions in neural network models trained on various datasets.
result Influence estimates are often erroneous for deeper networks and require regularization.

Aims to optimize influence spread in social networks using bandit algorithms.

problem Maximizing influence spread in unknown social networks.
method Combines Thompson Sampling and Epsilon Greedy algorithms with automatic ensemble learning.
result Demonstrates effectiveness of automatic ensemble learning for combinatorial bandit problems.