AMPLab Publications

• CYCLADES: Conflict-free Asynchronous Machine Learning
• A Variational Perspective on Accelerated Methods in Optimization
• A Linearly-Convergent Stochastic L-BFGS Algorithm
• SparkNet: Training Deep Networks on Spark
• Perturbed Iterate Analysis for Asynchronous Stochastic Optimization
• Parallel Correlation Clustering on Big Graphs
• Stale View Cleaning: Getting Fresh Answers from Stale Materialized Views
• Automating Model Search for Large Scale Machine Learning
• Distributed Estimation of Generalized Matrix Rank: Efficient Algorithms and Lower Bounds
• A General Analysis of the Convergence of ADMM
• Adding vs. Averaging in Distributed Primal-Dual Optimization
• Spectral methods meet EM: A provably optimal algorithm for crowdsourcing
• Changepoint Analysis for Efficient Variant Calling
• SMASH: A Benchmarking Toolkit for Human Genome Variant Calling
• The missing piece in complex analytics: Low latency, scalable model management and serving with Velox
• Communication-Efficient Distributed Dual Coordinate Ascent
• On the Convergence Rate of Decomposable Submodular Function Minimization
• Parallel Double Greedy Submodular Maximization
• Scaling Up Crowd-Sourcing to Very Large Datasets: A Case for Active Learning
• Estimation, Optimization, and Parallelism when Data is Sparse
• Knowing When You’re Wrong: Building Fast and Reliable Approximate Query Processing Systems
• Lower Bounds on the Performance of Polynomial-Time Algorithms for Sparse Linear Regression
• Distributed Low-rank Subspace Segmentation
• Streaming Variational Bayes
• On Statistics, Computation and Scalability
• Computational and Statistical Tradeoffs via Convex Relaxation
• MLI: An API for Distributed Machine Learning
• Optimistic Concurrency Control for Distributed Unsupervised Learning
• MAD-Bayes: MAP-based Asymptotic Derivations from Bayes
• Small-Variance Asymptotics for Exponential Family Dirichlet Process Mixture Models
• Privacy Aware Learning
• Revisiting K-Means: New algorithms via Bayesian Nonparametrics
• A General Bootstrap Performance Diagnostic
• MLbase: A Distributed Machine-learning System
• Mixed Membership Matrix Factorization
• Combinatorial Clustering and the Beta Negative Binomial Process
• Matrix Concentration Inequalities via the Method of Exchangeable Pairs
• A Scalable Bootstrap for Massive Data
• Ergodic Subgradient Descent
• Bayesian Bias Mitigation for Crowdsourcing
• Divide-and-Conquer Matrix Factorization
• The SCADS Director: Scaling a Distributed Storage System Under Stringent Performance Requirements
• Managing Data Transfers in Computer Clusters with Orchestra