Package uk.ac.liv.auction.agent

Classes defining trading agents and strategies

See:
Description

Interface Summary

AdaptiveStrategy	Strategies implementing this interface indicate that they are based on a learning algorithm.
FixedQuantityStrategy	Strategies implementing this interface indicate that they bid a constant quantity in each auction round.
Strategy	Classes implementing this interface define trading strategies for round-robin traders.
TradingAgent	Classes implementing this interface can trade in round-robin auctions, as implemented by the RoundRobinAuction class.
ValuationPolicy	A commodity valuation policy for RoundRobinTrader agents.

Class Summary

AbstractStrategy	An abstract implementation of the Strategy interface that provides skeleton functionality for making trading decisions.
AbstractTradingAgent	An abstract class representing a simple agent trading in a round-robin auction.
AdaptiveStrategyImpl
AgentGroup	A class representing an arbitrary grouping of agents.
BuyerIntervalValuer	Buyers configured with this valuation policy will receive a unique private value from a common set of values starting at minValue and incrementing by step as each agent is assigned a valuation.
DailyRandomValuer	A valuation policy in which we are allocated a new random valuation at the end of each day.
DiscreteLearnerStrategy	A class representing a strategy in which we adapt our bids using a discrete learning algorithm.
EquilibriumPriceStrategy	A strategy which will bid at the true equilibrium price, if profitable, or bid truthfully otherwise.
FixedPriceStrategy
FixedQuantityStrategyImpl	An abstract implementation of FixedQuantityStrategy.
FixedValuer	A valuation policy in which we maintain a fixed private valuation independent of time or auction.
GDStrategy	An implementation of the Gjerstad Dickhaut strategy.
IntervalValuer	Agents configured with this valuation policy will receive a unique private value from a common set of values starting at minValue and incrementing by step as each agent is assigned a valuation at agent setup time.
KaplanStrategy	An implementation of Todd Kaplan's sniping strategy.
MarkupStrategyDecorator	This strategy decorates a component strategy by bidding a fixed proportional markup over the price specified by the underlying component strategy.
MDPStrategy	A trading strategy that uses an MDP learning algorithm, such as the Q-learning algorithm, to adapt its trading behaviour in successive auction rounds.
MixedStrategy	A class representing a mixed strategy.
MomentumStrategy
PriestVanTolStrategy
ProportionalMarkupStrategy	This strategy bids at the specified percentage markup over the agent's current valuation.
PureSimpleStrategy	A trading strategy in which we bid a constant mark-up on the agent's private value.
RandomConstrainedStrategy	A trading strategy that in which we bid a different random markup on our agent's private value in each auction round.
RandomScheduleValuer	A valuation policy which specifies a randomly-generated series of valuations for each unit of commodity.
RandomUnconstrainedStrategy	A trading strategy in which an agent bid regardless its private value.
RandomValuer	A valuation policy in which we randomly determine our valuation across all auctions and all units at agent-initialisation time.
SellerIntervalValuer	Sellers configured with this valuation policy will receive a unique private value from a common set of values starting at minValue and incrementing by step as each agent is assigned a valuation.
SimpleMomentumStrategy
StimuliResponseStrategy	A trading strategy that uses a stimuli-response learning algorithm, such as the Roth-Erev algorithm, to adapt its trading behaviour in successive auction rounds by using the agent's profits in the last round as a reward signal.
TruthTellingStrategy

Package uk.ac.liv.auction.agent Description

Classes defining trading agents and strategies