Research Project: Large scale machine learning with MALT
At NECLA, I am working on MALT, a machine learning library that integrates with existing machine learning software (including NECLA MiLDE
) and provides data parallel machine learning. MALT provides abstractions for fine-grained in-memory updates using one-sided RDMA, limiting data movement costs during incremental model updates. MALT allows machine learning developers to specify the dataflow and apply communication and representation optimizations. Currently, MALT can be used to provide data-parallelism to existing ML applications written in C++ and Lua and based on SVM, matrix factorization and neural networks. MALT provides fault tolerance, network efficiency and speedup to ML applications.