The paper categorizes four types of scale-up: smart, dumb, forced, and fumbled.
problem Growing ventures in size and maintaining efficiency.
method Identifying modularity and speed as key factors, categorizing four types of scale-up.
result Modularity and speed are crucial for successful scale-up.
ADVGP scales up Gaussian process regression to large datasets efficiently.
problem Expensive computational cost of traditional GP inference for large datasets.
method Asynchronous Distributed Variational Gaussian Process (ADVGP) with weight space augmentation and asynchronous proximal gradient optimization.
result ADVGP achieves superior prediction accuracy for large-scale regression tasks.
This paper solves parameter estimation with ordered ℓ 2 \ell_2 ℓ 2 regularization using ADMM.
problem Scaling up ordered ℓ 2 \ell_2 ℓ 2 regularization for large-scale data. method Alternating Direction Method of Multipliers (ADMM) for ordered ℓ 2 \ell_2 ℓ 2 regularization. result ADMM-O ℓ 2 \ell_{2} ℓ 2 outperforms or matches state-of-the-art methods in parameter estimation. Two approaches scale up DNN optimization for diverse edge devices.
problem Optimizing DNNs for edge devices with varying performance requirements.
method Reuse performance predictors on proxy devices and build scalable predictors.
result Optimized DNN designs for many different edge devices without lengthy optimization.
This thesis examines the accuracy of scaling VaR estimates for longer holding periods.
problem The accuracy of VaR estimates for longer holding periods using the square root of time rule.
method Examined VaR scaling for longer holding periods using empirical analysis.
result Scaling can provide good estimates of VaR but may lead to significant losses over time.
Estimates network size and community sizes from a random sample.
problem Estimating the size of large networks and their communities.
method PULSE algorithm for efficient population size estimation.
result PULSE accurately estimates network and community sizes.
Scaling up model and data size improves imitation learning in single-agent games.
problem Limited recovery of expert behavior in single-agent games using imitation learning.
method Investigate the effect of scaling model and data size on imitation learning performance.
result IL loss and mean return scale with compute budget, resulting in power laws.
Active testing for large language models is made more efficient and accurate.
problem Efficient evaluation of large language models with limited labels.
method Cost-saving measures and in-context learning for constructing a surrogate model.
result Significantly more accurate evaluations of LLM performance compared to random data acquisition.
GaDei optimizes deep learning training-as-a-service by reducing communication traffic.
problem High communication traffic in deep learning training-as-a-service.
method GaDei is an optimized shared-memory scale-up parameter server design.
result GaDei significantly outperforms state-of-the-art parameter-server based implementations.
New method for matrix completion with row and column similarities.
problem Matrix completion with row and column similarities.
method Iterative model selection with Hutchinson estimator.
result Effective model selection for optimal smoothing parameters.
New method estimates order book dynamics efficiently.
problem Understanding complex order book dynamics in financial markets.
method Nonparametric estimation of branching ratio matrix for multivariate Hawkes process.
result Reveals relationships between order book events.
Method scales up ML science by measuring multiple molecules at once.
problem Scaling up ML-driven science with wet lab experiments.
method Neural extension of compressed sensing for function space.
result Proves orders-of-magnitude gains in information density.
New method improves kernel approximation for larger datasets.
problem Efficiently approximate kernel functions for large datasets.
method Monte Carlo integration for numerical approximation of kernel functions.
result Improved convergence behavior and empirical support for better kernel estimates.
A new R package for high-dimensional regression and precision matrix estimation.
problem High-dimensional linear regression and precision matrix estimation challenges.
method flare package implements various regression methods and extensions for sparse precision matrix estimation.
result The flare package is efficient and scalable for large problems.
Bayesian optimization speeds up bioprocess development across scales.
problem Costly and complex bioprocess development across scales and biocatalyst selection.
method Multi-fidelity batch Bayesian optimization framework integrating Gaussian Processes and mixed-variable optimization.
result Reduction in experimental costs and increased yield in bioprocess optimization.
Deep learning improves image reconstruction, but scaling up training sets doesn't significantly boost performance.
problem Understanding the impact of training set size on deep learning image reconstruction.
method Empirical study and analytical characterization of performance scaling laws.
result Scaling up training set size does not significantly improve reconstruction quality for deep learning.
Paper improves tree probability estimation using stochastic optimization and variance reduction.
problem Improving tree probability estimation in phylogenetic inference.
method Introduces computationally efficient methods for training SBNs and variance reduction for optimization.
result Methods outperform previous baseline methods in tree topology probability estimation and Bayesian phylogenetic inference.
We consider the problem of clustering a set of high-dimensional data points into sets of low-dimensional linear subspaces. The number of subspaces, their dimensions, and their orientations are unknown. We propose a simple and low-complexity clustering algorithm based on thresholding the correlations between the data po…
Random features are improved by variance-reducing couplings, enhancing machine learning models.
problem Improving the efficiency and accuracy of random features in machine learning.
method Using optimal transport theory to find couplings that reduce variance in random features.
result Theoretical and practical gains in efficiency and accuracy for various machine learning models.
We consider large-scale Markov decision processes (MDPs) with parameter uncertainty, under the robust MDP paradigm. Previous studies showed that robust MDPs, based on a minimax approach to handle uncertainty, can be solved using dynamic programming for small to medium sized problems. However, due to the "curse of dimen…
R package huge simplifies graph estimation for high-dimensional data.
problem Estimating high-dimensional undirected graphs from data.
method Uses recent results in literature, including recent graph estimation methods.
result Improves on existing package glasso by providing more features and better efficiency.
New method quantifies uncertainty in distributed regression.
problem Large datasets make traditional regression techniques ineffective.
method Data-driven approach to uncertainty quantification for averaged estimator.
result Rigorous theoretical guarantees for sup-norm consistency.
Asynchronous SVI scales up Bayesian computation for massive data.
problem Scaling stochastic variational inference to massive data.
method Lock-free asynchronous parallel implementation of SVI.
result Linear speed-up and asymptotic convergence rate O ( 1 / ( T ) ) O(1/\sqrt(T)) O ( 1/ ( T )) . HCNAF models complex conditional distributions for probabilistic occupancy forecasting.
problem Modeling complex conditional probability density functions for occupancy forecasting.
method Hyper-Conditioned Neural Autoregressive Flow (HCNAF) combining AF and hyper-network.
result HCNAF achieves state-of-the-art performance in self-driving datasets.
Noisy natural gradient improves variational inference for Bayesian neural nets.
problem Tradeoff between simple and complex variational families in Bayesian neural nets.
method Adaptive weight noise in natural gradient ascent to implicitly fit variational posteriors.
result Noisy natural gradient algorithms can train full-covariance variational posteriors efficiently.
New method speeds up sparse graph neural networks training on dense hardware.
problem Training sparse graph neural networks is slow on custom hardware.
method Inspired by sparse matrix optimization, developed techniques for dense hardware.
result Sparse graph neural networks trained in 13 minutes on 512-core TPUv2 Pod.
Paper proposes a new PLL framework with a progressive identification algorithm.
problem Weakly supervised learning with partial labels.
method Flexible model and optimization algorithm for PLL, progressive identification algorithm.
result Established an estimation error bound and set new state of the art.
Machine learning methods are used to discover complex nonlinear relationships in biological and medical data. However, sophisticated learning models are computationally unfeasible for data with millions of features. Here we introduce the first feature selection method for nonlinear learning problems that can scale up t…
Scaling multinomial logistic regression to datasets with very large number of data points and classes is challenging. This is primarily because one needs to compute the log-partition function on every data point. This makes distributing the computation hard. In this paper, we present a distributed stochastic gradient d…
We present a generic framework for parallel coordinate descent (CD) algorithms that includes, as special cases, the original sequential algorithms Cyclic CD and Stochastic CD, as well as the recent parallel Shotgun algorithm. We introduce two novel parallel algorithms that are also special cases---Thread-Greedy CD and …
BELIEF method efficiently selects features from Big Data sets.
problem Feature selection for large datasets in Big Data.
method Distributed feature weighting algorithm using redundancy elimination.
result BELIEF method ranks millions of features in parallel efficiently.
Dynamic topic models (DTMs) are very effective in discovering topics and capturing their evolution trends in time series data. To do posterior inference of DTMs, existing methods are all batch algorithms that scan the full dataset before each update of the model and make inexact variational approximations with mean-fie…
GA method reduces gradient staleness in cloud computing.
problem Gradient staleness in asynchronous SGD methods.
method Gap-Aware (GA) method that penalizes stale gradients linearly to the Gap.
result GA outperforms existing methods in final test accuracy.
This paper develops a bootstrap method to estimate errors in Random Fourier Features.
problem Inability to estimate the error of Random Fourier Features approximations.
method Develops a bootstrap approach to numerically estimate the errors of RFF approximations.
result Specific, flexible, and adaptive error estimates for RFF approximations.
We present a scalable Bayesian model for low-rank factorization of massive tensors with binary observations. The proposed model has the following key properties: (1) in contrast to the models based on the logistic or probit likelihood, using a zero-truncated Poisson likelihood for binary data allows our model to scale …
New methods for skill rating in sports using state-space models.
problem Improving skill rating in competitive sports.
method State-space models, sequential Monte Carlo, discrete hidden Markov models.
result Advantages of state-space models for time-varying player skills.
A new Gaussian process regression method infers implicit manifold structure from data.
problem Scaling Gaussian process regression to high-dimensional data.
method Proposes a fully differentiable Gaussian process regression technique that infers implicit manifold structure from data.
result Improves predictive performance and calibration of standard Gaussian process regression in high-dimensional settings.
A parallel algorithm improves gLasso scalability for large datasets.
problem Scalability issues in existing gLasso algorithms for large-scale datasets.
method Divide-and-conquer based parallel algorithm (DC-gLasso) with two iterations.
result Proven to recover true model under certain conditions and improve time efficiency.
Stochastic variational inference (SVI) lets us scale up Bayesian computation to massive data. It uses stochastic optimization to fit a variational distribution, following easy-to-compute noisy natural gradients. As with most traditional stochastic optimization methods, SVI takes precautions to use unbiased stochastic g…
Expands small recommendation datasets to industrial scale.
problem Disconnection between academic and industrial data scales.
method Randomized fractal expansions using Kronecker Graph Theory.
result Generated synthetic data sets with 1.2B ratings, 2.2M users, and 855K items.
One approach for constructing copula functions is by multiplication. Given that products of cumulative distribution functions (CDFs) are also CDFs, an adjustment to this multiplication will result in a copula model, as discussed by Liebscher (J Mult Analysis, 2008). Parameterizing models via products of CDFs has some a…
New algorithm reduces communication in distributed eigenspace estimation.
problem Efficiently estimating eigenspaces in distributed settings without excessive communication.
method Communication-efficient distributed algorithm using Procrustean alignment.
result Achieves similar error rate to centralized estimator for PCA.
New method decomposes sensory information from neurons into specific stimuli and features.
problem Understanding how much and what specific information neurons encode.
method Introduced axioms for meaningful stimulus-wise decomposition and derived a tractable solution using diffusion models.
result Can efficiently estimate contributions of specific stimuli and features to encoded information.
Proposes FMC for fair clustering with independent parameters.
problem Finding clusters with balanced sensitive attribute proportions.
method Model-based clustering using finite mixture model with mini-batch learning.
result FMC scales up easily and can handle non-metric data.
Analyzes generalization error in distributed linear regression.
problem Understanding generalization performance in distributed learning.
method Analytical characterization of generalization error in linear regression with distributed learning.
result Generalization error increases dramatically when nodes estimate close to the number of observations.
Study shows scaling up models doesn't always improve downstream tasks.
problem Understanding why scaling up models doesn't always improve downstream performance.
method Systematic study of 4800 experiments on various models, analyzing performance on 20 downstream tasks.
result Performance on downstream tasks saturates as model size increases, revealing a nonlinear relationship.
New insights into the top-K sparse softmax gating function for deep learning.
problem Understanding the theoretical effects of the top-K sparse softmax gating function on density and parameter estimations.
method Using a Gaussian mixture of experts, novel loss functions, and theoretical analysis.
result The convergence rates of density and parameter estimations are parametric under certain conditions, but slow under over-specified models.
A faster method for density estimation using denoising score matching with random Fourier features.
problem Intractability of normalizing constant in density estimation.
method Denoising Score Matching with Random Fourier Features.
result The method is computationally faster and scalable to complex high-dimensional data.