Zeno: Distributed Stochastic Gradient Descent with Suspicion-based Fault-tolerance.
Cong XieSanmi KoyejoIndranil GuptaPublished in: ICML (2019)
Keyphrases
- fault tolerance
- fault tolerant
- distributed systems
- stochastic gradient descent
- distributed computing
- peer to peer
- mobile agents
- group communication
- single point of failure
- database replication
- fault management
- load balancing
- response time
- step size
- replicated databases
- loss function
- matrix factorization
- least squares
- failure recovery
- text categorization
- random forests
- multiple kernel learning
- data streams
- training data
- data sets