Overfitting frequently occurs in deep learning. In this paper, we propose a novel regularization method called Drop-Activation to reduce overfitting and improve generalization. The key idea is to drop nonlinear activation functions by setting them to be identity functions randomly during training time. During testing, …
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.
168,695 papers · 148 categories
Trend · papers per month
48 results for “performance drop”
Deep neural networks have dramatically achieved great success on a variety of challenging tasks. However, most successful DNNs have an extremely complex structure, leading to extensive research on model compression.As a significant area of progress in model compression, traditional gradual pruning approaches involve an…
Deep neural networks (DNNs) have been proven to have many redundancies. Hence, many efforts have been made to compress DNNs. However, the existing model compression methods treat all the input samples equally while ignoring the fact that the difficulties of various input samples being correctly classified are different…
Out-of-distribution (OOD) detection approaches usually present special requirements (e.g., hyperparameter validation, collection of outlier data) and produce side effects (e.g., classification accuracy drop, slower energy-inefficient inferences). We argue that these issues are a consequence of the SoftMax loss anisotro…
New metrics reveal oversmoothing in GNNs more accurately than traditional methods.
problem Oversmoothing in graph neural networks reduces model performance.
method Rank-based metrics to measure oversmoothing in GNNs.
result Rank-based metrics consistently capture oversmoothing, while energy-based metrics often fail.
Chunking is a significant CL problem, accounting for half of performance drop, and current methods don't address it.
problem Chunking of data in continual learning.
method Analyzing and addressing the chunking sub-problem in continual learning.
result Current CL algorithms perform poorly on chunking, only as well as plain SGD training when there is no distribution shift.
Round balls minimize liquid drop model volumes ≤ 1.
problem Minimizing volumes in liquid drop models.
method Proved uniqueness of minimizers for small volumes.
result Round balls uniquely minimize volumes ≤ 1.
Machine learning is currently dominated by largely experimental work focused on improvements in a few key tasks. However, the impressive accuracy numbers of the best performing models are questionable because the same test sets have been used to select these models for multiple years now. To understand the danger of ov…
Empirical study shows GANs overfit and drop modes when training is deterministic.
problem Understanding overfitting and mode drop in GAN training.
method Empirical analysis of GAN training with and without stochasticity.
result GANs overfit and drop modes when training is deterministic.
The study examines mass drop and multiplicity in mean curvature flow.
problem Analyzing mass drop and multiplicity in mean curvature flow.
method Defined Brakke flow with variational inequality, proved mass drop conditions.
result Mass drop and multiplicity one conjecture are equivalent for Brakke flows.
Deep neural networks are vulnerable against adversarial examples. In this paper, we propose to train and test the networks with randomly subsampled images with high drop rates. We show that this approach significantly improves robustness against adversarial examples in all cases of bounded L0, L2 and L_inf perturbation…
This paper analyzes the configurations of shapes that shows a spacelike liquid drop in Minkowski space deposited over a spacelike plane . We assume the presence of a uniform gravity field directed toward and that the volume of the drop is prescribed. Our interest are the liquid drops that are critical points of …
System predicts HIV patients at risk of dropping out of care.
problem High non-adherence and dropout rates among HIV patients.
method Predictive machine learning model based on resource constraints, stability, and fairness.
result Model performs 3x better than baseline for clinical use and 2.3x better for city-wide use.
This study assesses the impact of non-IID data in federated learning, revealing significant performance drops.
problem The impact of non-IID data on federated learning model performance.
method Empirical analysis using Hellinger Distance to measure distribution differences, benchmarking four strategies for handling non-IID data.
result Significant performance drops occur at specific HD thresholds, especially for extreme non-IIDness.
Dropping a tiny fraction of preferences can significantly alter the rankings of top LLMs.
problem Robustness of LLM ranking systems to small changes in preference data.
method A computational method based on the Bradley-Terry model to evaluate robustness.
result Top LLM rankings can be highly sensitive to the removal of a small fraction of preferences.
When training clinical prediction models from electronic health records (EHRs), a key concern should be a model's ability to sustain performance over time when deployed, even as care practices, database systems, and population demographics evolve. Due to de-identification requirements, however, current experimental pra…
Study addresses RTB model performance drops due to distribution shifts.
problem Distribution shifts between training and target environments in RTB markets.
method Applies Exponential Tilt Reweighting Alignment (ExTRA) algorithm to estimate and correct model weights.
result Demonstrates improved RTB model performance using ExTRA algorithm.
Dropout has proven to be an effective technique for regularization and preventing the co-adaptation of neurons in deep neural networks (DNN). It randomly drops units with a probability during the training stage of DNN. Dropout also provides a way of approximately combining exponentially many different neural networ…
Paper resolves Huisken's conjecture without strict genus drop theorem.
problem Huisken's genericity conjecture in mean curvature flow in R^3.
method Short density-drop theorem + Bamler-Kleiner multiplicity-one theorem for tangent flows.
result Fully resolves Huisken's conjecture without strict genus drop theorem.
Study on liquidation games with market drop-out, proving unique equilibria.
problem Analyzing portfolio liquidation with market drop-out constraints.
method Proves existence and uniqueness of equilibria using integral equations.
result Existence and uniqueness of equilibria in both mean-field and finite-player games.
Drop-Muon updates only some layers, speeding up training.
problem Conventional deep learning optimizers update all layers at once, which can be inefficient.
method Drop-Muon updates only a subset of layers per step, with randomized schedules.
result Drop-Muon achieves up to 1.4x faster training time with similar accuracy.
Deep learning models struggle with new data in stock price trend prediction.
problem Stock price trend prediction using Deep Learning models.
method Examination of fifteen state-of-the-art DL models on LOB data, using LOBCAST framework.
result All models show significant performance drop with new data, questioning their market applicability.
RDIS fills missing values in time series data explicitly.
problem Missing values in time series data.
method Random Drop Imputation with Self-training.
result RDIS achieves competitive results on real-world datasets.
The Frank-Wolfe (FW) algorithm has been widely used in solving nuclear norm constrained problems, since it does not require projections. However, FW often yields high rank intermediate iterates, which can be very expensive in time and space costs for large problems. To address this issue, we propose a rank-drop method …
Under-parameterization hinders deep RL's efficiency.
problem Implicit under-parameterization impairs data-efficiency in deep RL.
method Characterized and mitigated the rank collapse of value network features.
result Controlling rank collapse improves deep RL performance.
The past few years have witnessed the fast development of different regularization methods for deep learning models such as fully-connected deep neural networks (DNNs) and Convolutional Neural Networks (CNNs). Most of previous methods mainly consider to drop features from input data and hidden layers, such as Dropout, …
Study connects hyperbolic geometry to membrane shapes.
problem Understanding the shapes of biological membranes.
method Relates geometry of hyperbolic space to Helfrich model.
result Establishes a connection between membrane shapes and hyperbolic geometry.
Price Drops, Fluctuations, and Correlation in a Multi-Agent Model of Stock Marketscond-mat.stat-mech
In this paper we compare market price fluctuations with the response to fundamental price drops within the Lux-Marchesi model which is able to reproduce the most important stylized facts of real market data. Major differences can be observed between the decay of spontaneous fluctuations and of changes due to external p…
Cross-Representation Transferability of Adversarial Attacks: From Spectrograms to Audio Waveformscs.SD
This paper shows the susceptibility of spectrogram-based audio classifiers to adversarial attacks and the transferability of such attacks to audio waveforms. Some commonly used adversarial attacks to images have been applied to Mel-frequency and short-time Fourier transform spectrograms, and such perturbed spectrograms…
Select-DC reduces GFLOPS for uncertainty estimation in neural networks.
problem Computational inefficiency in estimating model uncertainty for low-latency applications.
method Select-DC uses a subset of layers to model epistemic uncertainty with MCDC, reducing GFLOPS.
result Significant reduction in GFLOPS required for uncertainty estimation with marginal performance loss.
Method diagnoses model performance under distribution shifts.
problem Understanding and improving model performance under distribution shifts.
method DIstribution Shift DEcomposition (DISDE) method to attribute performance drop to distribution shifts.
result Shows how model performance can be improved across different distribution shifts.
This work introduces a method to attribute model performance drops to distribution shifts.
problem Attributing performance drops of machine learning models to distribution shifts.
method Formulated as a cooperative game, value of a set of distributions is defined as the change in model performance when only that set of distributions changes. Importance weighting method for computing the value of an arbitrary set of distributions is derived. Quantifying the contribution of each distribution as its Shapley value.
result Demonstrated the effectiveness of the method on various case studies.
Paper proposes a method to maintain ASR performance on new tasks without forgetting old ones.
problem Mitigating forgetting in ASR models when learning new tasks.
method A novel explainability-based knowledge distillation combined with response-based knowledge distillation.
result Our method outperforms existing ones in mitigating forgetting on multi-stage sequential training tasks.
Google uses continuous streams of data from industry partners in order to deliver accurate results to users. Unexpected drops in traffic can be an indication of an underlying issue and may be an early warning that remedial action may be necessary. Detecting such drops is non-trivial because streams are variable and noi…
Lower Ricci curvature bound prevents first Betti number from dropping more than dimension in collapsing manifolds.
problem Understanding how the first Betti number behaves under manifold collapse with Ricci curvature bounds.
method Analyzing sequences of Riemannian manifolds with lower Ricci curvature bounds.
result The first Betti number cannot drop more than the dimension in collapsing manifolds.
Improves performance of deep GCNs by controlling node feature variance.
problem Performance degradation in deep Graph Convolutional Networks (GCNs).
method Experimentally examined the role of TRANs and PROPs in GCNs, introduced Node Normalization (NodeNorm).
result Node Normalization effectively controls node feature variance, improving GCN performance in deep models.
Enhances time-series modeling by dropping patches, improving efficiency and adaptability.
problem Improving masked time-series modeling efficiency and adaptability.
method DropPatch method, which randomly drops sub-sequence level patches.
result DropPatch improves pre-training efficiency and adaptability in various learning scenarios.
CODA uses a new dropout technique inspired by constructivism learning to improve deep learning performance.
problem Existing dropout methods fail to differentiate among instances, leading to overfitting.
method CODA incorporates structural information and uses a Bayesian nonparametric method to create a better dropout technique.
result CODA outperformed other state-of-the-art dropout techniques on 5 real-world datasets.
Study reveals statistical bias in dataset replication, reducing accuracy drop from 11-14% to 3.6%.
problem Statistical bias in dataset replication affects model generalization accuracy.
method Analyzed ImageNet-v2, identified and corrected for bias, and compared results.
result Correcting bias reduces accuracy drop from 11-14% to 3.6%.
A new loss function improves neural networks' out-of-distribution detection without side effects.
problem Neural networks struggle with out-of-distribution detection due to SoftMax loss issues.
method Proposes IsoMax loss replacing SoftMax loss, maintaining high entropy and fast inferences.
result Significantly improves neural networks' out-of-distribution detection performance.
The study of heavy-tailed distributions in economic and financial systems has been widely addressed since financial time series has become a research subject.After the eighties, several "highly improbable" market drops were observed (e.g. the 1987 stock market drop known as "Black Monday" and on even more recent ones, …
Deep neural networks achieve state-of-the-art results on several tasks while increasing in complexity. It has been shown that neural networks can be pruned during training by imposing sparsity inducing regularizers. In this paper, we investigate two techniques for group-wise pruning during training in order to improve …
Dimension drop of harmonic measure for some finite range random walks on Fuchsian Schottky groupsmath.GT
Random walks on Fuchsian Schottky groups have harmonic measures with lower dimension.
problem Understanding the dimensionality of harmonic measures for random walks.
method Analyzing finite range random walks on Fuchsian Schottky groups.
result Harmonic measures have dimension strictly less than the limit set's Hausdorff dimension.
Study on stability of 3D sessile drops, identifying degenerate kernel.
problem Linear stability of three-dimensional sessile drops with a free contact line.
method Derived constrained second variation, formulated Jacobi problem, combined geometric and Fourier analysis.
result Kernel of the constrained Jacobi operator is exactly the space of horizontal translations under pressure-volume nondegeneracy.
Study evaluates how well question-answering models generalize to new data types.
problem Generalization of question-answering models to new data types.
method Constructed new test sets from different domains and evaluated models' performance.
result Models show significant performance drops when tested on new data types.
Many tasks in natural language understanding require learning relationships between two sequences for various tasks such as natural language inference, paraphrasing and entailment. These aforementioned tasks are similar in nature, yet they are often modeled individually. Knowledge transfer can be effective for closely …
We study the long time existence theory for a non local flow associated to a free boundary problem for a trapped non liquid drop. The drop has free boundary components on two horizontal plates and its free energy is anisotropic and axially symmetric. For axially symmetric initial surfaces with sufficiently large volume…
Dropout improves MIL performance on noisy WSI classification.
problem Noisy feature embeddings and weak supervision in MIL for WSI classification.
method Empirical exploration of dropout in MIL, proposing MIL-Dropout.
result MIL-Dropout boosts MIL performance with minimal computational cost.