arXiv 2014

19,071 papers
In 2014, research on arXiv comprised 982 papers with a focus on scalable optimization and Bayesian inference. Key research directions included the development of stochastic gradient MCMC for scalable Bayesian inference and neural transport methods for multimodal probability landscapes. There was also an emphasis on decentralized optimization, particularly through consensus methods and the Alternating Direction Method of Multipliers (ADMM), often applied to dynamic network topologies and large power grids. Statistical research during this period addressed high-dimensional singular covariance, non-parametric detection of non-linear dependencies, and heavy-tailed stochastic optimization. Prolific authors such as Giles Hooker, Günther Uhlmann, and Claire Tomlin contributed to these fields. Sample works explored diverse applications including low-rank matrix recovery, computer model calibration, and the analysis of temporal patterns in WiFi-interactive social populations. The year's output reflected a trend toward integrating physical constraints into decentralized energy markets and parallelizing sequential computations for modern hardware.

Top Research Directions

Research directions most represented by arXiv 2014 papers.

The Empirical Validation Gap 60
Overcoming the Convexity Constraint in Modern Optimization 46
Semidefinite Relaxation for Non-Convex Engineering Optimization 39
Generalizing Provable Control to Nonlinear Systems 39
Dynamic Preference Modeling in Interactive Systems 39
From Fractional Solutions to Discrete Decisions 35
Incremental Maintenance of Dynamic Graph Indices 35
The Scalability Bottleneck of Singular Value Decomposition 35
Scaling Gaussian Processes Beyond Cubic Complexity 35
Overcoming the Cubic Complexity of Kernel Methods 35
Greedy Maximization for Submodular Subset Selection 33
Overlapping Community Detection in Complex Networks 32
Stationary Point Convergence in Nonconvex Optimization 31
Kernel Hyperparameter Sensitivity in Distributional Metrics 31
Variational Inference as an Optimization Problem 29

Top Authors

AuthorPapers
Alfred O. Hero38
Yoshua Bengio23
Mohamed-Slim Alouini22
Chunhua Shen19
Joseph Halpern19
Uwe Aickelin18
Ping Li17
Svetha Venkatesh16
Anshumali Shrivastava15
Jitendra Malik15
Truyen Tran14
Aram Galstyan13
Bei Yu13
David Z. Pan13
Krishnendu Chatterjee13
Lenka Zdeborova13
Shuicheng YAN13
Aaditya Ramdas12
Aaron Roth12
Anastasios Kyrillidis12

Impactful Papers

Papers from arXiv 2014 that shaped research clusters.

Foundational

Influential

Papers

Title
A two-step learning approach for solving full and almost full cold start problems in dyadic prediction
Audio Source Separation with Discriminative Scattering Networks
Audio Surveillance: a Systematic Review
Audit Games with Multiple Defender Resources
Augmentative Message Passing for Traveling Salesman Problem and Graph Partitioning
Augmented Neural Networks for Modelling Consumer Indebtness
Augmenting recommendation systems using a model of semantically-related terms extracted from user behavior
A unified approach for multi-object triangulation, tracking and camera calibration
A unified framework for thermal face recognition
A unified framework for the study of the PLS estimator's properties
A Unified Semantic Embedding: Relating Taxonomies and Attributes
A Unified View of Large-scale Zero-sum Equilibrium Computation
A Unifying Framework in Vector-valued Reproducing Kernel Hilbert Spaces for Manifold Regularization and Co-Regularized Multi-view Learning
A Unifying Hierarchy of Valuations with Complements and Substitutes
A universal law in human mobility

« prev Page 65/382 next »