The combination of the re-parameterization trick with the use of variational auto-encoders has caused a sensation in Bayesian deep learning, allowing the training of realistic generative models of images and has considerably increased our ability to use scalable latent variable models. The re-parameterization trick is …
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
24 results for “re-parameterization”
ResRep prunes CNNs without losing accuracy by separating remembering and forgetting.
problem Pruning CNNs to reduce FLOPs without sacrificing accuracy.
method Decoupling remembering and forgetting in CNNs, using SGD for remembering and a novel update rule for forgetting.
result Achieved lossless pruning with high compression ratio (76.15% accuracy on ImageNet with 45% FLOPs reduction).
Quantum advantage in derivative pricing requires 8k qubits and 54M T-depth.
problem Quantum advantage in pricing derivatives.
method Re-parameterization method combining pre-trained variational circuits and fault-tolerant quantum computing.
result Benchmark use cases require 8k logical qubits and a T-depth of 54 million.
This thesis presents some geometric insights into three different types of two player prediction games -- namely general learning task, prediction with expert advice, and online convex optimization. These games differ in the nature of the opponent (stochastic, adversarial, or intermediate), the order of the players' mo…
The expectation-maximization (EM) algorithm is an iterative method for finding maximum likelihood estimates when data are incomplete or are treated as being incomplete. The EM algorithm and its variants are commonly used for parameter estimation in applications of mixture models for clustering and classification. This …
Safe, Scalable, and Accurate Bayes Posterior Sampling for Large-Data Generalized Linear Mixed Modelsstat.ME
New method for Bayesian inference on large datasets.
problem Scalable sampling for Bayesian generalized linear mixed models on large datasets.
method Mirror Langevin dynamics with data subsampling, post-processing for variance estimation.
result Asymptotic, order-wise correct estimation of posterior variance.
Federated Learning is a distributed learning paradigm with two key challenges that differentiate it from traditional distributed optimization: (1) significant variability in terms of the systems characteristics on each device in the network (systems heterogeneity), and (2) non-identically distributed data across the ne…
The paper analyzes the amplitude of functions on the sphere, improving FDA methods.
problem Analyzing trajectories on non-linear manifolds with time variability.
method Developed tools for temporal alignment, geodesic computation, and mean calculation on .
result Efficient and accurate tools for analyzing manifold-valued functions on .
The paper proposes a new method for creating interpretable models using convex optimization.
problem Creating models that are both accurate and interpretable for decision-making.
method Formulates convex learning problems that combine interpretability with accuracy, using operator theory and parametric nonlinear models.
result Shows how to create efficient surrogate models that are both accurate and interpretable.
Quaternion self-attention reduces computational cost and improves performance.
problem Existing quaternion self-attention increases computational cost and diverges attention distributions.
method Proposes a shared-score quaternion self-attention mechanism.
result Reduces score-computation multiplications by 75% and softmax operations from four to one.
New method samples Jeffreys prior for objective Bayesian inference.
problem Sampling from Jeffreys prior is challenging.
method Metropolis-Adjusted Langevin Algorithm
result Samples can be directly used in Bayesian methods.
The goal of online prediction with expert advice is to find a decision strategy which will perform almost as well as the best expert in a given pool of experts, on any sequence of outcomes. This problem has been widely studied and and regret bounds can be achieved for convex losses (\cite{zin…
New framework analyzes deep neural networks using feature probabilities.
problem Degenerate situation in over-parameterized DNNs.
method Mean-field framework representing DNNs by feature probabilities and functions.
result Global convergence proof for over-parameterized Res-Net training.
AdaptOn achieves logarithmic regret in adaptive control of unknown partially observable linear systems.
problem Adaptive control in partially observable linear dynamical systems.
method AdaptOn algorithm that estimates system dynamics through online learning and gradient descent.
result AdaptOn achieves a logarithmic regret bound of polylog(T) after T steps.
Global information is essential for dense prediction problems, whose goal is to compute a discrete or continuous label for each pixel in the images. Traditional convolutional layers in neural networks, initially designed for image classification, are restrictive in these problems since the filter size limits their rece…
Improved VI method for deep mixed models in finance.
problem Inaccurate and slow variational inference in high dimensions.
method Natural gradient hybrid VI method targeting joint posterior.
result Natural gradient method is faster and more accurate than existing methods.
Bayesian optimization improved for high-dimensional outputs using randomized priors.
problem Efficient global optimization of high-dimensional black-box functions.
method Deep learning framework with bootstrapped ensembles of neural architectures with randomized priors.
result Superior performance in tasks with high-dimensional outputs compared to state-of-the-art methods.
Efficiently learns exponential family distributions with i.i.d. samples.
problem Learning natural parameters of truncated exponential families efficiently.
method Proposes a novel loss function and computationally efficient estimator.
result Achieves optimal sample complexity and asymptotic normality.
New method for initializing low-rank neural networks improves performance.
problem Training low-rank neural networks efficiently and accurately.
method Inspired by function approximation, proposes a novel low-rank initialization framework.
result Demonstrates significant gap between spectral and low-rank initialization approaches.
We simplify debiasing of LASSO estimates for practical computation.
problem Bias in LASSO estimates for high-dimensional data.
method Reformulate optimization problem to compute debiasing matrix W directly.
result Efficient closed-form solution for W with unique optimal solution.
The paper introduces a measure to assess the relative value of a delta-Symmetric Strangle under the Black-Scholes model.
problem Measuring the relative value of a delta-Symmetric Strangle under the Black-Scholes model.
method Developed a new measure of relative value in terms of delta and volatility, bounded by a simple function of delta.
result The relative value of a delta-Symmetric Strangle is bounded by a simple function of delta and is independent of other factors.
Researchers develop a new spatial process model for non-Gaussian data.
problem Non-Gaussian spatial data with asymmetry and heavy-tailedness.
method Re-parameterized Unified Skew-Normal (SUN) distribution, GSUN process, neural Bayes inference with GATs.
result GSUN process captures non-Gaussian spatial data properties and outperforms conventional models.
We propose a new proximal, path-following framework for a class of constrained convex problems. We consider settings where the nonlinear---and possibly non-smooth---objective part is endowed with a proximity operator, and the constraint set is equipped with a self-concordant barrier. Our approach relies on the followin…
RAE improves image representation learning with simplified design choices.
problem Improving image representation learning using pretrained vision encoders.
method Generalized RAE formulation, complementary working mechanisms of RAE and REPA, and free CFG guidance.
result RAEv2 achieves state-of-the-art results with 10x faster convergence and less training time.