Kernel embeddings of distributions and the Maximum Mean Discrepancy (MMD), the resulting distance between distributions, are useful tools for fully nonparametric two-sample testing and learning on distributions. However, it is rarely that all possible differences between samples are of interest -- discovered difference…
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.
168,657 papers · 148 categories
Trend · papers per month
48 results for “distributional difference”
New metrics improve regression evaluation across different data distributions.
problem Difficulty in comparing regression evaluations across datasets with varying distributions.
method Modification of regression metrics by weighting with the inverse distribution of function values or samples using a Gaussian kernel density estimator.
result New metrics are less sensitive to changing distributions, especially when correcting by the marginal distribution in .
Distributed machine learning is an approach allowing different parties to learn a model over all data sets without disclosing their own data. In this paper, we propose a weighted distributed differential privacy (WD-DP) empirical risk minimization (ERM) method to train a model in distributed setting, considering differ…
ABROCA assesses algorithmic bias, revealing skewed distributions that inflate results.
problem Detecting nuanced performance differences in classifier fairness.
method Study of ABROCA metric's statistical properties under various conditions.
result ABROCA distributions are skewed, inflating results by chance in imbalanced classes.
Paper characterizes DLN distribution, its properties, and estimation methods.
problem No specific problem stated, focuses on DLN distribution properties.
method Characterization of PDF, CDF, moments; generalization to N-dimensions; methods to handle double-exponential nature.
result Characterization of DLN distribution and its properties, including estimation methods.
This work examines the sensitivity of energy distance to mean differences compared to covariance differences.
problem The sensitivity of energy distance to mean differences compared to covariance differences when distributions are close.
method Analyzes the energy distance in the case where distributions are close, focusing on sensitivity to mean and covariance differences.
result Energy distance is more sensitive to mean differences than covariance differences when distributions are close.
Principal Differences Analysis: Interpretable Characterization of Differences between Distributionsstat.ML
We introduce principal differences analysis (PDA) for analyzing differences between high-dimensional distributions. The method operates by finding the projection that maximizes the Wasserstein divergence between the resulting univariate populations. Relying on the Cramer-Wold device, it requires no assumptions about th…
Paper proposes a new method to aggregate multiple sources with different label distributions.
problem Aggregating from multiple target-shifted sources with different label distributions.
method Unified framework to select relevant sources for domain adaptation with limited label, unsupervised, and label partial unsupervised scenarios.
result Empirical results significantly outperform baselines.
Analyzed US firm data 1970-2019, identifying scale effects and distributional forms.
problem Understanding differences between small and large firms over time.
method Examined all public US firms, used stylized facts and DLN distribution analysis.
result Small firms are systematically different from large firms, with scale-dependent heteroskedasticity.
SAPAG attacks distributed learning by reconstructing true training data from gradients.
problem Privacy attacks on distributed learning systems through gradients.
method SAPAG uses a Gaussian kernel-based gradient difference distance measure.
result SAPAG can reconstruct training data on various DNNs and at different training phases.
This work proposes a new method to match distributions across different spaces using cycle-consistent maps.
problem Matching distributions across different spaces with consistent bidirectional maps.
method A novel unbalanced Monge optimal transport formulation for matching distributions on different spaces, employing cycle-consistent maps.
result The proposed discrepancy captures the cycle-consistent GAN framework and provides theoretical support.
New algorithms improve distributional TD learning with linear approximations.
problem Estimating return distributions in reinforcement learning.
method Fine-grained analysis of linear-categorical Bellman equation, variance reduction techniques.
result Tight sample complexity bounds for distributional TD learning with linear approximations.
Quantile TD learning outperforms classical TD learning for value estimation.
problem Temporal-difference learning in reinforcement learning.
method Quantile Temporal-Difference Learning (QTD) for policy evaluation.
result QTD offers superior performance to classical TD learning, even in tabular settings.
Implied and Realized Volatility: A Study of Distributions and the Distribution of Differenceq-fin.ST
We study distributions of realized variance (squared realized volatility) and squared implied volatility, as represented by VIX and VXO indices. We find that Generalized Beta distribution provide the best fits. These fits are much more accurate for realized variance than for squared VIX and VXO -- possibly another indi…
Distributed learning adapts to diverse devices, improving performance.
problem Training neural networks on devices with varying capabilities and resources.
method Each device trains a customized neural network, sharing parameters with others.
result Achieves higher rewards on more powerful devices without sacrificing weaker ones.
Transfer learning aims to learn robust classifiers for the target domain by leveraging knowledge from a source domain. Since the source and the target domains are usually from different distributions, existing methods mainly focus on adapting the cross-domain marginal or conditional distributions. However, in real appl…
An important goal common to domain adaptation and causal inference is to make accurate predictions when the distributions for the source (or training) domain(s) and target (or test) domain(s) differ. In many cases, these different distributions can be modeled as different contexts of a single underlying system, in whic…
Deep networks adapt to function regularity and data distribution.
problem Understanding deep learning's adaptability to function regularity and data distribution.
method Developed nonparametric approximation and estimation theories for a broad class of functions using deep ReLU networks.
result Deep neural networks are adaptive to different regularity of functions and nonuniform data distributions.
Are two sets of observations drawn from the same distribution? This problem is a two-sample test. Kernel methods lead to many appealing properties. Indeed state-of-the-art approaches use the distance between kernel-based distribution representatives to derive their test statistics. Here, we show that distan…
In open set learning, a model must be able to generalize to novel classes when it encounters a sample that does not belong to any of the classes it has seen before. Open set learning poses a realistic learning scenario that is receiving growing attention. Existing studies on open set learning mainly focused on detectin…
The analysis of the USA 2001 income distribution shows that it can be described by at least two main components, which obey the generalized Tsallis statistics with different values of the q parameter. Theoretical calculations using the gas kinetics model with a distributed saving propensity factor and two ensembles rep…
Training on mixed distributions improves test performance even when components are unrelated.
problem Improving test performance with mismatched training and test distributions.
method Analyzing mixture distributions with different training and test proportions.
result Distribution shift can be beneficial, improving test performance even when components are unrelated.
The study improves model performance prediction on unseen distributions.
problem Improving model performance prediction on unseen distributions.
method Connecting domain adaptation and predictive uncertainty techniques, investigating distributional distances and DoC.
result Difference of confidences (DoC) successfully estimates classifier performance change over various distribution shifts.
Functional brain networks are well described and estimated from data with Gaussian Graphical Models (GGMs), e.g. using sparse inverse covariance estimators. Comparing functional connectivity of subjects in two populations calls for comparing these estimated GGMs. Our goal is to identify differences in GGMs known to hav…
MaxRM uses random forests to minimize maximum risk across different environments.
problem Designing methods that generalize better to test environments with different distributions.
method Introducing variants of random forests based on the principle of MaxRM (Maximum Risk Minimization).
result Proved statistical consistency for the proposed method and provided an out-of-sample guarantee for MaxRM with regret.
In this paper, we present the results of Monte Carlo simulations for two popular techniques of long-range correlations detection - classical and modified rescaled range analyses. A focus is put on an effect of different distributional properties on an ability of the methods to efficiently distinguish between short and …
The Origin of Fat Tailsq-fin.GN
We propose a random walk model of asset returns where the parameters depend on market stress. Stress is measured by, e.g., the value of an implied volatility index. We show that model parameters including standard deviations and correlations can be estimated robustly and that all distributions are approximately normal.…
New algorithm for RL using mean embeddings of return distributions.
problem Improving reinforcement learning algorithms for dynamic programming.
method Mean embeddings of return distributions, novel algorithms for RL.
result Asymptotic convergence and improved performance in deep RL.
Exponential distribution is ubiquitous in the framework of multi-agent systems. An alternative approach with an economic motivation to derive the exponential distribution in the framework of iterations in the space of distributions is disclosed.
Kernel Tests of Equivalencestat.ML
Kernel tests assess equivalence between distributions without assuming specific moments.
problem Traditional goodness-of-fit tests fail to detect meaningful distributional differences.
method Proposes kernel-based tests using kernel Stein discrepancy and Maximum Mean Discrepancy.
result Tests assess the absence of meaningful distributional differences under controlled error rates.
Domain generalization is the problem of machine learning when the training data and the test data come from different data domains. We present a simple theoretical model of learning to generalize across domains in which there is a meta-distribution over data distributions, and those data distributions may even have dif…
This work analyzes how multi-agent reinforcement learning can bridge the gap to reality in distributed multi-robot systems.
problem Collaborative learning in distributed multi-robot systems with varying sensors and actuators.
method Simulation-based analysis using PPO and Bullet physics engine, considering different types of perturbations.
result PPO's robustness is affected by the presence of different types of perturbations and the number of agents experiencing them.
An analytic solution for asset allocation with Laplace distribution.
problem Asset allocation with multivariate Laplace distribution.
method Specialization of elliptically symmetric distribution theory to Laplace distribution, accounting for dimensionality and variance rescaling.
result A result consistent with conjecture but with differences due to omitted term and rescaling.
A new measure scales MMD to assess distribution closeness.
problem Testing statistical significance of distribution closeness.
method Norm-adaptive MMD (NAMMD) for distributional discrepancy.
result NAMMD-based DCT has higher test power than MMD-based DCT.
Recent interest in the external validity of prediction models (i.e., the problem of different train and test distributions, known as dataset shift) has produced many methods for finding predictive distributions that are invariant to dataset shifts and can be used for prediction in new, unseen environments. However, the…
In this paper we perform a statistical analysis of the high-frequency returns of the IBEX35 Madrid stock exchange index. We find that its probability distribution seems to be stable over different time scales, a stylized fact observed in many different financial time series. However, an in-depth analysis of the data us…
Analyzes financial return distributions over various time scales.
problem Understanding the changing nature of financial return distributions over time.
method Modeling return distributions using power-law, stretched exponential, and q-Gaussian functions.
result The 'inverse-cubic power-law' is still a good fit for short-term returns, but market dynamics are more complex.
Survey of performative prediction, a machine learning setup causing distribution shifts.
problem Machine learning models causing shifts in the environment they predict.
method Classification of performative prediction settings based on distribution map information.
result Introduction of new solution concepts and theoretical analyses.
Efficient bandit exploration for various distributions without distribution-specific tuning.
problem Optimizing exploration in multi-armed bandit models for different distributions.
method Sub-sampling Duelling Algorithms (SDA) with Random Block sampling for efficient exploration.
result Achieves asymptotically optimal regret for Bernoulli, Gaussian, and Poisson distributions.
Correlation between Risk Aversion and Wealth distributioncond-mat.stat-mech
Different models of capital exchange among economic agents have been proposed recently trying to explain the emergence of Pareto's wealth power law distribution. One important factor to be considered is the existence of risk aversion. In this paper we study a model where agents posses different levels of risk aversion,…
We propose methods for distributed graph-based multi-task learning that are based on weighted averaging of messages from other machines. Uniform averaging or diminishing stepsize in these methods would yield consensus (single task) learning. We show how simply skewing the averaging weights or controlling the stepsize a…
Recently, Mike and Farmer have constructed a very powerful and realistic behavioral model to mimick the dynamic process of stock price formation based on the empirical regularities of order placement and cancelation in a purely order-driven market, which can successfully reproduce the whole distribution of returns, not…
Since their introduction a year ago, distributional approaches to reinforcement learning (distributional RL) have produced strong results relative to the standard approach which models expected values (expected RL). However, aside from convergence guarantees, there have been few theoretical results investigating the re…
We devise a distributional variant of gradient temporal-difference (TD) learning. Distributional reinforcement learning has been demonstrated to outperform the regular one in the recent study \citep{bellemare2017distributional}. In the policy evaluation setting, we design two new algorithms called distributional GTD2 a…
Properties of distributions of the number of trades in different intraday time intervals for five stocks traded in MICEX are studied. The dependence of the mean number of trades on the capital turnover is analyzed. Correlation analysis using factorial and moments demonstrates the multifractal nature of these dist…
We propose a \textbf{uni}fied \textbf{f}ramework for \textbf{i}mplicit \textbf{ge}nerative \textbf{m}odeling (UnifiGem) with theoretical guarantees by integrating approaches from optimal transport, numerical ODE, density-ratio (density-difference) estimation and deep neural networks. First, the problem of implicit gene…
Study on continuous sequence classification with distribution uncertainty.
problem Classifying continuous sequences with varying distribution uncertainty.
method Proposes distribution-free tests for three test designs: fixed-length, sequential, and two-phase tests.
result Error probabilities decay exponentially fast for all test designs.
How fair is an equitable distribution?physics.soc-ph
Envy is a rather complex and irrational emotion. In general, it is very difficult to obtain a measure of this feeling, but in an economical context envy becomes an observable which can be measured. When various individuals compare their possessions, envy arises due to the inequality of their different allocations of co…