Peter Vamplew John Yearwood Richard Dazeley Adam Berry On the Limitations of Scalarisation for Multi-objective Reinforcement Learning of Pareto Fronts. 372-378 2008 Australasian Conference on Artificial Intelligence http://dx.doi.org/10.1007/978-3-540-89378-3_37 conf/ausai/2008 db/conf/ausai/ausai2008.html#VamplewYDB08