Online Learning with Vector Costs and Bandits with Knapsacks.
Thomas KesselheimSahil SinglaPublished in: COLT (2020)
Keyphrases
- online learning
- regret bounds
- e learning
- knapsack problem
- online course
- distance education
- computer mediated
- total cost
- feature vectors
- blended learning
- higher education
- distance learning
- multi armed bandits
- reinforcement learning
- training data
- cost savings
- online algorithms
- feature extraction
- online education
- stochastic systems
- corporate training
- decision trees