Incremental learning system

It is possible to interrupt a learning process part way through if more data become available or if some data are revised The GCS is, therefore, well suited to the development of incremental or dynamic learning systems. 

Over the past decade we have explored a number of phosphine ligands for incremental improvements of present systems and for studies in new types of reductions and hydroformylations. Even though bisphosphines continue to dominate the field, we have found fairly efficient monophosphines. This chapter deals with these and other systems that we have explored for a-phenylacrylic acid, dehydrovaline, and hydroformylation of vinyl acetate. Even though we have not solved efficiently any of these cases, we ve learned a lot about the behavior of a variety of ligands and hope that this knowledge will contribute to these and other unsolved problems in this field. 

The constant system parameters used by the rulebase had to be learned before it could be run satisfactorily. Upper and lower carbon monoxide optimum limits and maximum and minimum increments for the air inlet valves all needed adjusting. 

---