On parallelizing SGD for pairwise learning to rank in collaborative filtering recommender systems

Keyphrases

• Stochastic Gradient Descent 100%
• Collaborative Recommender Systems 100%
• Pairwise Loss Function 100%
• Learning to Rank 100%
• Parallelizing 100%
• Pairwise Learning to Rank 100%
• Matrix Factorization 50%
• Pointwise Learning 50%
• Lock-free 50%
• Parallel Stochastic Gradient Descent 50%
• Shared Memory 50%
• Block Partition 50%
• Web-scale 50%
• Practical Algorithm 50%

Computer Science

• Collaborative Filtering 100%
• Recommender Systems 100%
• Gradient Descent 100%
• Speed-up 33%
• Shared Memory 33%
• Block Partitioning 33%
• Experimental Result 33%
• Matrix Factorization 33%