SSMs have a built-in bias towards low-frequency components, which can be adjusted.
problem Frequency bias in SSMs affects their performance on long-range sequences.
method Proposed two mechanisms to tune frequency bias: scaling initialization or applying a Sobolev-norm-based filter.
result Tuning frequency bias improves SSMs' performance on long-range sequence learning tasks.
Convolutional GANs favor low spatial frequencies, affecting fine detail generation.
problem Understanding GANs' limitations in high spatial frequency learning.
method Proposed method to manipulate GANs' bias against high spatial frequencies.
result Convolutional GANs have a bias against learning high spatial frequencies.
Local convolutions bias neural networks towards high-frequency adversarial examples.
problem High-frequency adversarial examples in neural networks.
method Analysis of different linear and nonlinear architectures, focusing on the impact of local convolution operations.
result Local convolutions induce an implicit bias towards high frequency features, leading to high-frequency adversarial examples.
Study reveals how neural network biases align with adversarial attack frequencies.
problem Correlation between neural network biases and adversarial attacks.
method Fourier transform analysis of network implicit bias and adversarial perturbations.
result Network bias and adversarial attack frequencies are highly correlated.
Study shows neural networks learn low frequencies first, proposing solutions.
problem Frequency bias in neural network learning process.
method Developed a PDE to unravel frequency dynamics, used Fourier Features model.
result Appropriate weight initialization can eliminate or control frequency bias.
Frequency bias affects neural network training on non-uniform data.
problem Understanding how frequency bias impacts neural networks trained on non-uniformly distributed data.
method Used the Neural Tangent Kernel (NTK) model to explore the effect of variable density on training dynamics.
result Convergence time for learning a pure harmonic function depends on the local density at a point.
We study the relationship between the frequency of a function and the speed at which a neural network learns it. We build on recent results that show that the dynamics of overparameterized neural networks trained with gradient descent can be well approximated by a linear system. When normalized training data is uniform…
Deep neural networks can generalize by reducing high-frequency noise over time, not always following a monotonic learning bias.
problem Understanding the learning dynamics and generalization of over-parameterized DNNs.
method Experimental analysis of deep double descent, focusing on the spectral bias of DNNs.
result The high-frequency components of DNNs diminish over training, leading to a second descent in test error.
One of the fundamental questions of cultural evolutionary research is how individual-level processes scale up to generate population-level patterns. Previous studies in music have revealed that frequency-based bias (e.g. conformity and novelty) drives large-scale cultural diversity in different ways across domains and …
This paper proposes a novel multiscale estimator for the integrated volatility of an Ito process, in the presence of market microstructure noise (observation error). The multiscale structure of the observed process is represented frequency-by-frequency and the concept of the multiscale ratio is introduced to quantify t…
High-frequency trading models fail due to overfitting and survivor bias.
problem Failure of hybrid DRL-EC trading systems in high-frequency environments.
method Deployed a population of 500 agents in a high-frequency cryptocurrency environment, analyzing failure modes through multi-disciplinary lens.
result Increasing model complexity without information asymmetry exacerbates systemic fragility.
FreSh shifts model's initial frequency spectrum to match target signal, improving neural representation performance.
problem MLPs' low-frequency bias limits capturing high-frequency details accurately.
method FreSh selects embedding hyperparameters to align model's initial output spectrum with target signal's spectrum.
result FreSh improves performance across various neural representation methods and tasks with minimal computational overhead.
We analyze realized volatilities constructed using high-frequency stock data on the Tokyo Stock Exchange. In order to avoid non-trading hours issue in volatility calculations we define two realized volatilities calculated separately in the two trading sessions of the Tokyo Stock Exchange, i.e. morning and afternoon ses…
FFN addresses spectral bias in neural value approximation, improving reinforcement learning performance.
problem Spectral bias in neural value approximation, leading to slow convergence and poor performance.
method Proposes Fourier feature networks (FFN) to overcome spectral bias by using a composite neural tangent kernel.
result FFN achieves state-of-the-art performance on challenging continuous control domains with faster convergence and better stability.
Safe-FinRL uses DRL for high-frequency stock trading, reducing bias and variance.
problem Challenges in applying DRL to high-frequency stock trading, especially bias and variance issues.
method Safe-FinRL separates financial time series into near-stationary short environments and uses Trace-SAC with a general retrace operator.
result Safe-FinRL reduces bias and variance significantly in near-stationary financial environments.
CNNs show sensitivity to low-frequency signals due to image frequency distribution.
problem Understanding why CNNs are sensitive to low-frequency signals.
method Theoretical analysis of CNN representations in frequency space.
result CNNs sensitivity to low-frequency signals is due to the frequency distribution of natural images.
FreST Loss decorrelates spatio-temporal dependencies in graph signals.
problem Complex spatio-temporal dependencies in graph-structured signals are not well captured by standard forecasting models.
method FreST Loss extends supervision to the joint spatio-temporal spectrum using Joint Fourier Transform (JFT).
result FreST Loss reduces estimation bias and improves forecasting accuracy on real-world datasets.
New method constrains CNN filter frequencies to improve robustness.
problem CNN bias towards low frequency components, leading to poor performance in scenario transformations.
method Frequency domain regularization by constraining filter spectra, training valid frequency range end-to-end.
result Demonstrated effectiveness in defending adversarial perturbations, reducing generalization gap, and improving transfer learning.
SKI speeds up Toeplitz Neural Networks by avoiding explicit decay bias and using frequency response.
problem Efficiently compute and update Toeplitz matrices in neural networks.
method Sparse plus low-rank decomposition, asymmetric SKI, frequency response modeling.
result Achieved significant speedup with minimal performance loss.
It remains a puzzle that why deep neural networks (DNNs), with more parameters than samples, often generalize well. An attempt of understanding this puzzle is to discover implicit biases underlying the training process of DNNs, such as the Frequency Principle (F-Principle), i.e., DNNs often fit target functions from lo…
sgboost reduces variable selection bias in boosting with balanced group selection.
problem Reduces variable selection bias in boosting algorithms.
method Simulation-based approach to balance selection frequencies of base-learners.
result Demonstrates efficacy through simulations and flexible group variable selection.
New Fourier-based diffusion model improves high-frequency generation quality.
problem Diffusion models struggle with high-frequency details.
method Analyzed and modified the forward process in Fourier space to equalize noise corruption across frequencies.
result Improved generation quality for high-frequency components.
Efficiently scales continuous kernels with sparse Fourier domain learning.
problem High computational and memory demands, spectral bias in continuous kernels.
method Sparse learning in the Fourier domain.
result Efficient scaling of continuous kernels, reduced computational and memory requirements, mitigated spectral bias.
We study the training process of Deep Neural Networks (DNNs) from the Fourier analysis perspective. We demonstrate a very universal Frequency Principle (F-Principle) -- DNNs often fit target functions from low to high frequencies -- on high-dimensional benchmark datasets such as MNIST/CIFAR10 and deep neural networks s…
FreDF improves forecasting by learning in the frequency domain.
problem Label autocorrelation in future forecasts is often overlooked in time series modeling.
method FreDF learns to forecast in the frequency domain to mitigate label autocorrelation.
result FreDF significantly outperforms existing methods in forecasting accuracy.
When estimating high-frequency covariance (quadratic covariation) of two arbitrary assets observed asynchronously, simple assumptions, such as independence, are usually imposed on the relationship between the prices process and the observation times. In this paper, we introduce a general endogenous two-dimensional nonp…
We develop a general multivariate aggregation property which encompasses the distinct versions of the property that were introduced by Neuberger [2012] and Bondarenko [2014] independently. This way, we classify new types of model-free realised characteristics for which risk premia may be estimated without bias. We focu…
Neural networks are known to be a class of highly expressive functions able to fit even random input-output mappings with 100% accuracy. In this work, we present properties of neural networks that complement this aspect of expressivity. By using tools from Fourier analysis, we show that deep ReLU networks are biased…
Measures of implied volatility roughness corrected for bias.
problem Bias in measuring implied volatility roughness.
method Examined implied volatility of short-term options and VIX index, corrected for bias.
result Corrected measures indicate appropriate proxies for underlying volatility.
This work analyzes how different layers in deep neural networks contribute to generalization error.
problem Understanding the role of each layer in deep neural networks for generalization.
method Spectral analysis, Neural Tangent Kernel, Hermite polynomials, Spherical Harmonics.
result Initial layers in deep neural networks have a larger bias towards high-frequency functions.
We show that gradient descent on full-width linear convolutional networks of depth L converges to a linear predictor related to the ℓ2/L bridge penalty in the frequency domain. This is in contrast to linearly fully connected networks, where gradient descent converges to the hard margin linear support vector m…
Physen-Noise2Noise tackles defocus deblurring in low-light conditions with physics-guided self-supervised learning.
problem Defocus deblurring in low-light conditions with complex biased noise.
method Physics-guided self-supervised deblurring framework that leverages noisy multi-frame observations and a learnable noise bias parameter.
result Physen-Noise2Noise consistently outperforms state-of-the-art methods in defocus deblurring with complex biased noise.
Develops a GMM method to estimate roughness in stochastic volatility models.
problem Estimating roughness in stochastic volatility models with fractional Brownian motion.
method GMM approach for log-normal models with integrated variance and noisy realized variance.
result Consistent and asymptotically normal parameter estimator with bias correction.
This work analyzes PINNs for advection-diffusion equations using NTK theory.
problem Understanding and resolving the training difficulties of PINNs for advection-diffusion equations.
method Neural Tangent Kernel (NTK) analysis of PINNs for the linear advection-diffusion equation (LAD).
result PINNs struggle due to spectral bias and convergence rate disparity, especially in advection-dominated and diffusion-dominated regimes.
When stock prices are observed at high frequencies, more information can be utilized in estimation of parameters of the price process. However, high-frequency data are contaminated by the market microstructure noise which causes significant bias in parameter estimation when not taken into account. We propose an estimat…
FAST selects coresets more efficiently by matching distributions in the frequency domain.
problem Efficiently selecting representative subsets of large datasets for deep learning.
method FAST uses spectral graph theory and CFD to match distributions, addressing limitations of existing methods.
result FAST significantly outperforms state-of-the-art coreset selection methods in accuracy and energy efficiency.
This work analyzes how often to update the target network in Q-learning.
problem Understanding the optimal frequency of target network updates in Q-learning.
method Formulated target updates as a nested optimization scheme, derived finite-time convergence analysis.
result Optimal target update frequency increases geometrically over time.
We introduce and show the existence of a Hawkes self-exciting point process with exponentially-decreasing kernel and where parameters are time-varying. The quantity of interest is defined as the integrated parameter T−1∫0Tθt∗dt, where θt∗ is the time-varying parameter, and we consider the high-frequency…
New holistic approach measures sample-level adversarial vulnerability for trustworthy systems.
problem Inherent bias in adversarial attacks across subgroups.
method Combining high-frequency feature reliance and sample-distance to decision boundary.
result Holistic approach improves adversarial vulnerability estimation and system trustworthiness.
The paper introduces a new price model based on entropy that better fits high-frequency market data.
problem Understanding fair prices in high-frequency markets with bid-ask imbalance.
method A parametrized family of prices derived from the Maximum Entropy Principle, minimizing bias given volume imbalance.
result The model can generate higher kurtosis and heavy-tailed distributions compared to standard models.
Estimates volatility of volatility and leverage effect using high-frequency options data.
problem Estimating volatility of volatility and leverage effect from high-frequency options data.
method Model-free estimators using characteristic function of price increments and spot volatility.
result Developed feasible inference methods for estimating volatility of volatility and leverage effect.
The Whittle likelihood is a widely used and computationally efficient pseudo-likelihood. However, it is known to produce biased parameter estimates for large classes of models. We propose a method for de-biasing Whittle estimates for second-order stationary stochastic processes. The de-biased Whittle likelihood can be …
New study shows how adversaries can bias fair machine learning models even with corrupted data.
problem Fairness concerns in machine learning models under data corruption.
method Study of fairness-aware learning algorithms under worst-case data manipulations.
result Natural learning algorithms optimizing for both accuracy and fairness are order-optimal in terms of corruption ratio and protected groups frequencies.
PRISM integrates diverse rewards in MORL, improving sample efficiency and Pareto coverage.
problem Heterogeneous MORL where dense objectives dominate, leading to poor sample efficiency.
method PRISM uses reflectional symmetry and ReSymNet to reconcile temporal-frequency mismatches and accelerate exploration.
result PRISM consistently outperforms sparse-reward baselines and oracles, achieving significant Pareto gains.
We examine the performance of six estimators of the power-law cross-correlations -- the detrended cross-correlation analysis, the detrending moving-average cross-correlation analysis, the height cross-correlation analysis, the averaged periodogram estimator, the cross-periodogram estimator and the local cross-Whittle e…
A hybrid model combines diffusion and neural operator methods for stress prediction in hyperelastic materials.
problem Challenges in predicting stress fields in hyperelastic materials with complex microstructures.
method A hybrid surrogate framework combining a conditional denoising diffusion probabilistic model (cDDPM) and a modified DeepONet.
result The hybrid model consistently outperforms traditional methods by one to two orders of magnitude.
A new method for multi-task learning improves performance without weakening inductive bias.
problem Joint optimization of parameters for multiple tasks remains challenging.
method Maximum Roaming, a novel parameter partitioning method inspired by dropout.
result Maximum Roaming improves performance compared to recent multi-task learning formulations.
CycleGAN-VC3 improves CycleGAN-VCs for mel-spectrogram conversion.
problem Ambiguity in CycleGAN-VC/VC2 effectiveness for mel-spectrogram conversion.
method Proposes CycleGAN-VC3 with time-frequency adaptive normalization (TFAN).
result CycleGAN-VC3 outperforms or matches CycleGAN-VC2 for mel-spectrogram conversion.