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.
Deeper neural networks learn lower frequency functions faster, according to a new principle.
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…
Previous studies have shown that deep neural networks (DNNs) with common settings often capture target functions from low to high frequency, which is called Frequency Principle (F-Principle). It has also been shown that F-Principle can provide an understanding to the often observed good generalization ability of DNNs. …
Why deep neural networks (DNNs) capable of overfitting often generalize well in practice is a mystery [#zhang2016understanding]. To find a potential mechanism, we focus on the study of implicit biases underlying the training process of DNNs. In this work, for both real and synthetic datasets, we empirically find that a…
Along with fruitful applications of Deep Neural Networks (DNNs) to realistic problems, recently, some empirical studies of DNNs reported a universal phenomenon of Frequency Principle (F-Principle): a DNN tends to learn a target function from low to high frequencies during the training. The F-Principle has been very use…
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…
New principle reveals how neural networks learn complex interactions.
Local convolutions bias neural networks towards high-frequency adversarial examples.
This research explores principles of Lipschitz continuity in neural networks for robustness and generalization.
Study shows neural networks learn low frequencies first, proposing solutions.
A new uncertainty principle helps traders better understand market activity.
This study proposes a trainable adaptive window switching (AWS) method and apply it to a deep-neural-network (DNN) for speech enhancement in the modified discrete cosine transform domain. Time-frequency (T-F) mask processing in the short-time Fourier transform (STFT)-domain is a typical speech enhancement method. To re…
Tumors often contain multiple subpopulations of cancerous cells defined by distinct somatic mutations. We describe a new method, PhyloWGS, that can be applied to WGS data from one or more tumor samples to reconstruct complete genotypes of these subpopulations based on variant allele frequencies (VAFs) of point mutation…
In this research, an emotion recognition system is developed based on valence/arousal model using electroencephalography (EEG) signals. EEG signals are decomposed into the gamma, beta, alpha and theta frequency bands using discrete wavelet transform (DWT), and spectral features are extracted from each frequency band. P…
Bayesian model reconstructs time and frequency data robustly.
We discuss the radiation problem of total reflection for a time-harmonic generalized Maxwell system in a non-smooth exterior domain with non-smooth inhomogeneous, anisotropic coefficients converging near infinity with a certain rate towards the identity. By means of the limiting absorption principle, a Fredholm alterna…
High-dimensional inference for sparse spectral precision matrices
Robustly detects jumps in high-frequency CIR and CKLS models.
Improves Bayesian optimization efficiency for mixed variable spaces.
This work analyzes how often to update the target network in Q-learning.
The paper introduces a new price model based on entropy that better fits high-frequency market data.
Convolutional neural networks (CNN) are widely used for speech emotion recognition (SER). In such cases, the short time fourier transform (STFT) spectrogram is the most popular choice for representing speech, which is fed as input to the CNN. However, the uncertainty principles of the short-time Fourier transform preve…
What return should you expect when you take on a given amount of risk? How should that return depend upon other people's behavior? What principles can you use to answer these questions? In this paper, we approach these topics by exploring the consequences of two simple hypotheses about risk. The first is a common-sense…
A nonlinear channel estimator using complex Least Square Support Vector Machines (LS-SVM) is proposed for pilot-aided OFDM system and applied to Long Term Evolution (LTE) downlink under high mobility conditions. The estimation algorithm makes use of the reference signals to estimate the total frequency response of the …
Spectral Independence Criterion helps infer cause-effect relationships in time series.
Study optimal liquidation strategies under partial information in high-frequency trading.
New method bypasses global fit for LISA's Galactic binaries, extracting population parameters directly.
Uncertainty principles such as Heisenberg's provide limits on the time-frequency concentration of a signal, and constitute an important theoretical tool for designing and evaluating linear signal transforms. Generalizations of such principles to the graph setting can inform dictionary design for graph signals, lead to …
Concerns about interpretability, computational resources, and principled inductive priors have motivated efforts to engineer sparse neural models for NLP tasks. If sparsity is important for NLP, might well-trained neural models naturally become roughly sparse? Using the Taxi-Euclidean norm to measure sparsity, we find …
Fidel-TS creates a new benchmark for time series forecasting models.
We consider an investor, whose portfolio consists of a single risky asset and a risk free asset, who wants to maximize his expected utility of the portfolio subject to managing the Value at Risk (VaR) assuming a heavy tailed distribution of the stock prices return. We use a stochastic maximum principle to formulate the…
In this work, we address the problem of musical timbre transfer, where the goal is to manipulate the timbre of a sound sample from one instrument to match another instrument while preserving other musical content, such as pitch, rhythm, and loudness. In principle, one could apply image-based style transfer techniques t…
Developed a diffusion model on spherical data, addressing geometric and stochastic challenges.
Alternative wavelet analysis method for financial signals.
Enhances CMS with Bayesian nonparametrics for better low-frequency token estimation.
We introduce the convolutional spectral kernel (CSK), a novel family of non-stationary, nonparametric covariance kernels for Gaussian process (GP) models, derived from the convolution between two imaginary radial basis functions. We present a principled framework to interpret CSK, as well as other deep probabilistic mo…
HyFAD improves time series imputation by combining time and frequency diffusion.
SSMs have a built-in bias towards low-frequency components, which can be adjusted.
Geometrically interprets frequency in electric circuits.
Archetypal analysis represents a set of observations as convex combinations of pure patterns, or archetypes. The original geometric formulation of finding archetypes by approximating the convex hull of the observations assumes them to be real valued. This, unfortunately, is not compatible with many practical situations…
Trading affects grid frequency fluctuations, making them more extreme.
This paper investigates Shampoo's heuristics and decouples preconditioner updates.
Study on frequencies of non-simple curves in surfaces of large genus.
CNNs show sensitivity to low-frequency signals due to image frequency distribution.
The paper examines how parabolic frequency behaves under Ricci flow and Ricci-harmonic flow on manifolds.
The paper analyzes a private likelihood-ratio test for frequency tables under differential privacy constraints.
New method constrains CNN filter frequencies to improve robustness.
Extracts causal brain dynamics across multiple scales.