# ENVIRONMENT

OSVERS=busternb

LOGBACK_VERSION=1.2.10

JPATH=lib/lazylav/ll.jar:lib/slf4j/logback-classic-${LOGBACK_VERSION}.jar:lib/slf4j/logback-core-${LOGBACK_VERSION}.jar:.

LLHASH=ceb1c5d1ca8109373d293b687fc55953fce5241d

# EXPERIMENT

DESIGNDATE=20230401
DATE=20230401
LABEL=20230505-MTOA
NAME=${LABEL}
PERFORMER="Andreas Kalaitzakis"
NBAGENTS=18
RATIO=0.1
NBCLSS=4
SRATIO=0.6
NBFEATURES=6
NBTASKS=3
MAX_ADAPTATION_RANK="0,1,2"
GAMES_SPECIALIZATION_RATIO=1.0
KNOWLEDGE_LIMIT=0.225
TASK_SELECTION_STRATEGY="BALANCED"
NBITERATIONS=80000
NBRUNS=20
RECORDMODE="allMeasures"
WINCONDITION="task"
ORG=godecisions
OP="oneCom"
IRRELEVANTFEATURES="2:3:4:5-2:3:4:5-2:3:4:5,2:3:4:5-0:1:4:5-0:1:2:3"
MERGES=300
SUBSTITUTIONS=300

OPT="-DmaxSubstitutions=${SUBSTITUTIONS} -DmaxMerges=${MERGES} -DallTasksIrrelevantFeatures=${IRRELEVANTFEATURES} -Dop=${OP} -Dorganiser=${ORG} -DknowledgeLimit=${KNOWLEDGE_LIMIT} -DtaskSelectionStrategy=${TASK_SELECTION_STRATEGY} -DspecializedGamesRatio=${GAMES_SPECIALIZATION_RATIO} -DmaxAdaptationRank=${MAX_ADAPTATION_RANK} -DnumberOfFeatures=${NBFEATURES} -Dratio=${RATIO} -DnumberOfAgents=${NBAGENTS} -DnumberOfIterations=${NBITERATIONS} -DnbRuns=${NBRUNS} -DnumberOfClasses=${NBCLSS} -DsampleRatio=${SRATIO} -DnumberOfTasks=${NBTASKS} -DrecordMode=${RECORDMODE} -DwinCondition=${WINCONDITION}"

DIRPREF=results

# DOCUMENTATION PARAMETERS

EXPE="Agents specialize by accepting or not to play a task."
#
HYPOTHESIS="Agents will improve their accuracy more on tasks they choose to play."
#
SETTING="Agents are trained with respect to different tasks and then coordinate upon acting on them. Each time they disagree, one agent adapts its knowledge with respect to the current task."

# DEFAULT VALUES
DESIGNER="Andreas Kalaitzakis"
EXPERIMENTER=${PERFORMER}
ANALYST=${PERFORMER}