Stochastic Optimization

Stochastic Optimization is the task of optimizing certain objective functional by generating and using stochastic random variables. Usually the Stochastic Optimization is an iterative process of generating random variables that progressively finds out the minima or the maxima of the objective functional. Stochastic Optimization is usually applied in the non-convex functional spaces where the usual deterministic optimization such as linear or quadratic programming or their variants cannot be used.

Source: ASOC: An Adaptive Parameter-free Stochastic Optimization Techinique for Continuous Variables

Papers

Recently Added Most Hyped Most Active Needs Verification Most Verified

Showing 1281–1290 of 1387 papers

Title	Date	Tasks	Status	Hype
Second-Order Stochastic Optimization for Machine Learning in Linear Time	Feb 12, 2016	BIG-bench Machine LearningSecond-order methods	CodeCode Available	1
Improved Dropout for Shallow and Deep Learning	Feb 6, 2016	Deep LearningStochastic Optimization	—Unverified	0
Variance-Reduced and Projection-Free Stochastic Optimization	Feb 5, 2016	Stochastic Optimization	—Unverified	0
A Kronecker-factored approximate Fisher matrix for convolution layers	Feb 3, 2016	Stochastic Optimization	CodeCode Available	0
On Deep Multi-View Representation Learning: Objectives and Optimization	Feb 2, 2016	Representation LearningStochastic Optimization	CodeCode Available	0
SCOPE: Scalable Composite Optimization for Learning on Spark	Jan 30, 2016	Stochastic Optimization	CodeCode Available	0
Variational Inference: A Review for Statisticians	Jan 4, 2016	Stochastic OptimizationVariational Inference	CodeCode Available	1
Bridging the Gap between Stochastic Gradient MCMC and Stochastic Optimization	Dec 25, 2015	Stochastic Optimization	CodeCode Available	0
Efficient Distributed SGD with Variance Reduction	Dec 9, 2015	Stochastic Optimization	—Unverified	0
Variance Reduction for Distributed Stochastic Gradient Descent	Dec 5, 2015	Stochastic Optimization	—Unverified	0

Show:10 25 50

← PrevPage 129 of 139Next →

All datasets CIFAR-100 WRN-28-10 - 200 Epochs CIFAR-10 WRN-28-10 - 200 Epochs CIFAR-10 ResNet-18 - 200 Epochs ImageNet ResNet-50 - 90 Epochs Penn Treebank (Character Level) 3x1000 LSTM - 500 Epochs CIFAR-10 CIFAR-100 ImageNet ResNet-50 - 50 Epochs ImageNet ResNet-50 - 60 Epochs AG News CoLA MNIST

Benchmark Results

#	Model	Metric	Claimed	Verified	Status
1	AvaGrad	Accuracy	81.24	—	Unverified
2	AdaShift	Accuracy	81.12	—	Unverified
3	Adam (eps-adjusted)	Accuracy	81.04	—	Unverified
4	SGD	Accuracy	80.95	—	Unverified
5	AdamW	Accuracy	79.87	—	Unverified
6	AdaBound	Accuracy	77.24	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Adam (eps-adjusted)	Accuracy	96.36	—	Unverified
2	AvaGrad	Accuracy	96.2	—	Unverified
3	SGD	Accuracy	96.14	—	Unverified
4	AdaShift	Accuracy	95.92	—	Unverified
5	AdamW	Accuracy	95.89	—	Unverified
6	AdaBound	Accuracy	94.6	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	SGD - cosine LR schedule	Accuracy	95.55	—	Unverified
2	Lookahead	Accuracy	95.27	—	Unverified
3	SGD	Accuracy	95.23	—	Unverified
4	ADAM	Accuracy	94.84	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	AvaGrad	Top 1 Accuracy	76.51	—	Unverified
2	SGD	Top 1 Accuracy	75.99	—	Unverified
3	AdamW	Top 1 Accuracy	72.9	—	Unverified
4	AdaBound	Top 1 Accuracy	72.01	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	AdaBound	Bit per Character (BPC)	2.86	—	Unverified
2	AdaShift	Bit per Character (BPC)	1.27	—	Unverified
3	AdamW	Bit per Character (BPC)	1.23	—	Unverified
4	AvaGrad	Bit per Character (BPC)	1.18	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Resnet18	Accuracy (max)	86.85	—	Unverified
2	Resnet34	Accuracy (max)	86.14	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Resnet18	Accuracy (max)	58.48	—	Unverified
2	Resnet34	Accuracy (max)	54.5	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	SGD	Top 5 Accuracy	92.15	—	Unverified
2	Lookahead	Top 1 Accuracy	75.13	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Lookahead	Top 1 Accuracy	75.49	—	Unverified
2	SGD	Top 1 Accuracy	75.15	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Bert	Accuracy (max)	93.99	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Bert	Accuracy (max)	86.34	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	MLP	NLL	0.05	—	Unverified