negmas.gb Package

Implements Generalized Bargaining Protocol (GB) set of mechanisms and basic negotiators.

Functions

all_negotiator_types()	Returns all the negotiator types defined in negmas.gb.negotiators
all_accept(responses)	rtype: Union[tuple, None, Literal['continue']]
any_accept(responses)	rtype: Union[tuple, None, Literal['continue']]

Classes

ResponseType(value[, names, module, ...])	Possible responses to offers during negotiation.
GBState([running, waiting, started, step, ...])
GBNMI(id, n_outcomes, outcome_space, ...[, ...])
ThreadState([new_offer, new_responses, ...])
NegotiatorMechanismInterface(id, n_outcomes, ...)	All information of a negotiation visible to negotiators.
GBComponent(*[, negotiator])
AcceptancePolicy(*[, negotiator])
OfferingPolicy(*[, negotiator])
ProposalPolicy	alias of OfferingPolicy
Model	alias of GBComponent
LimitedOutcomesAcceptancePolicy(prob[, ...])	Accepts from a list of predefined outcomes
NegotiatorAcceptancePolicy(acceptor, *[, ...])	Uses a negotiator as an offering strategy
ConcensusAcceptancePolicy(strategies, *[, ...])	Accepts based on concensus of multiple strategies
AllAcceptanceStrategies(strategies, *[, ...])	Accept only if all children accept, end only if all of them end, otherwise reject
AnyAcceptancePolicy(strategies, *[, negotiator])	Accept any children accept, end or reject only if all of them end or reject
AcceptImmediately(*[, negotiator])	Accepts immediately anything
RejectAlways(*[, negotiator])	Rejects everything
EndImmediately(*[, negotiator])	Rejects immediately anything
AcceptAbove(limit[, above_reserve, negotiator])	Accepts outcomes with utilities in the given top limit fraction above reserve/minimum (based on above_resrve ).
RandomAcceptancePolicy([p_acceptance, ...])
AcceptTop([fraction, k, negotiator])	Accepts outcomes that are in the given top fraction or top k.
AcceptBest([best_util, negotiator])	Accepts Only the best outcome.
TFTAcceptancePolicy(partner_ufun, recommender, *)	An acceptance strategy that concedes as much as the partner (or more)
ACNext(offering_strategy[, alpha, beta, ...])	Implements the ACnext acceptance strategy based on our next offer.
ACLast([alpha, beta, negotiator])	Implements the AClast acceptance strategy based on our last offer.
ACLastKReceived([k, alpha, beta, op, negotiator])	Accepts $omega$ if $lpha u(my-next-offer) + eta > f(u( ext{utils of offers received in the last k steps))$
ACLastFractionReceived([fraction, alpha, ...])	Accepts $omega$ if $lpha u(my-next-offer) + eta > f(u( ext{utils of offers received in the given fraction of time}))$
ACTime(tau, *[, negotiator])	Implements the ACtime acceptance strategy based on our next offer.
AcceptAfter	alias of ACTime
AcceptAround([relative_time, eps, negotiator])	Accepts around the given relative time (i.e. eps from it).
AcceptBetween(min[, max, negotiator])	Accepts in the given range of relative times.
ACConst([th, negotiator])	Accepts outcomes with utilities above the given threshold
AcceptAnyRational(*[, negotiator])	Accepts any rational outcome.
AcceptBetterRational([accepted, negotiator])	Accept first rational outcomes and then accept only outcomes better than the all accepted so far.
AcceptNotWorseRational([accepted, negotiator])	Accept any outcome not worse than the best so far.
CABOfferingPolicy([next_indx, sorter, ...])
WAROfferingPolicy([next_indx, sorter, ...])
LimitedOutcomesOfferingPolicy(outcomes[, ...])	Offers from a given list of outcomes
NegotiatorOfferingPolicy(*[, negotiator])	Uses a negotiator as an offering strategy
ConcensusOfferingPolicy(strategies, *[, ...])	Offers based on concensus of multiple strategies
RandomConcensusOfferingPolicy(strategies[, ...])	Offers a random response from the list of strategies (different strategy every time).
UnanimousConcensusOfferingPolicy(strategies, *)	Offers only if all offering strategies gave exactly the same outcome
UtilBasedConcensusOfferingPolicy(strategies, *)	Offers from the list of stratgies (different strategy every time) based on outcome utilities
MyBestConcensusOfferingPolicy(strategies, *)	Offers my best outcome from the list of stratgies (different strategy every time).
MyWorstConcensusOfferingPolicy(strategies, *)	Offers my worst outcome from the list of stratgies (different strategy every time) based on outcome utilities
NoneOfferingPolicy(*[, negotiator])	Always offers None which means it never gets an agreement.
RandomOfferingPolicy(*[, negotiator])	Always offers None which means it never gets an agreement.
OfferTop([fraction, k, negotiator])	Offers outcomes that are in the given top fraction or top k.
OfferBest([best, negotiator])	Offers Only the best outcome.
TFTOfferingPolicy(partner_ufun, recommender)	An acceptance strategy that concedes as much as the partner (or more)
MiCROOfferingPolicy([next_indx, sorter, ...])
TimeBasedOfferingStrategy([curve, ...])
ConcessionRecommender(*[, negotiator])	Decides the level of concession to use
KindConcessionRecommender([kindness, ...])	A simple recommender that does one small concession first then a tit-for-tat response
OfferSelectorProtocol(*args, **kwargs)	Can select the best offer in some sense from a list of offers based on an inverter
OfferSelector(*args, **kwargs)	Can select the best offer in some sense from a list of offers based on an inverter
RandomOfferSelector(*args, **kwargs)
BestOfferSelector(*args, **kwargs)
MedianOfferSelector(*args, **kwargs)
WorstOfferSelector(*args, **kwargs)
OfferOrientedSelector([distance_fun])	Selects the nearest outcome to the pivot outcome which is updated before responding
FirstOfferOrientedSelector([distance_fun])	Selects the offer nearest the partner's first offer
LastOfferOrientedSelector([distance_fun])	Selects the offer nearest the partner's last offer
BestOfferOrientedSelector([distance_fun])	Selects the offer nearest the partner's best offer for me so far
OutcomeSetOrientedSelector([distance_fun, ...])	Selects the nearest outcome to a set of pivot outcomes which is updated before responding
PartnerOffersOrientedSelector([...])	Orients offes toward the set of past opponent offers
MultiplicativePartnerOffersOrientedSelector([...])	Orients offes toward the set of past opponent offers.
AdditivePartnerOffersOrientedSelector(*args)	Orients offes toward the set of past opponent offers.
UtilityInverter(*args[, offer_selector])	A component that can recommend an outcome based on utility
UtilityBasedOutcomeSetRecommender([...])	Recommends a set of outcome appropriate for proposal
UFunModel(*[, negotiator])	A SAOComponent that can model the opponent's utility function.
FrequencyUFunModel(*[, negotiator])	A PartnerUfunModel that uses a simple frequency-based model of the opponent offers.
FrequencyLinearUFunModel(*[, negotiator])	A PartnerUfunModel that uses a simple frequency-based model of the opponent offers assuming the ufun is LinearAdditiveUtilityFunction .
ZeroSumModel([above_reserve, rank_only, ...])	Assumes a zero-sum negotiation (i.e. $u_o$ = $-u_s$ ).
GBMechanism(*args[, evaluator_type, ...])
ParallelGBMechanism(*args, **kwargs)
SerialGBMechanism(*args, **kwargs)
GBNegotiator([preferences, ufun, name, ...])	Base class for all GB negotiators.
LimitedOutcomesNegotiator([...])	A negotiation agent that uses a fixed set of outcomes in a single negotiation.
LimitedOutcomesAcceptor([...])	A negotiation agent that uses a fixed set of outcomes in a single negotiation.
ToughNegotiator([can_propose])	Accepts and proposes only the top offer (i.e. the one with highest utility).
TopFractionNegotiator([min_utility, ...])	Offers and accepts only one of the top outcomes for the negotiator.
UtilBasedNegotiator(*args[, stochastic, ...])	A negotiator that bases its decisions on the utility value of outcomes only.
TimeBasedNegotiator(*args[, offering_curve, ...])	Represents a time-based negotiation strategy that is independent of the offers received during the negotiation.
TimeBasedConcedingNegotiator(*args[, ...])	Represents a time-based negotiation strategy that is independent of the offers received during the negotiation.
BoulwareTBNegotiator(*args, **kwargs)	A Boulware time-based negotiator that conceeds sub-linearly
LinearTBNegotiator(*args, **kwargs)	A Boulware time-based negotiator that conceeds linearly
ConcederTBNegotiator(*args, **kwargs)	A Boulware time-based negotiator that conceeds super-linearly
AspirationNegotiator(*args[, ...])	Represents a time-based negotiation strategy that is independent of the offers received during the negotiation.
FirstOfferOrientedTBNegotiator(*args[, ...])	A time-based negotiator that selectes outcomes from the list allowed by the current utility level based on their utility value and how near they are to the partner's first offer
LastOfferOrientedTBNegotiator(*args[, ...])	A time-based negotiator that selectes outcomes from the list allowed by the current utility level based on their utility value and how near they are to the partner's last offer
BestOfferOrientedTBNegotiator(*args[, ...])	A time-based negotiator that selectes outcomes from the list allowed by the current utility level based on their utility value and how near they are to the partner's past offer with the highest utility for me
AdditiveParetoFollowingTBNegotiator(*args[, ...])	A time-based negotiator that selectes outcomes from the list allowed by the current utility level based on a weighted sum of their normalized utilities and distances to previous offers
MultiplicativeParetoFollowingTBNegotiator(*args)	A time-based negotiator that selectes outcomes from the list allowed by the current utility level based on a weighted sum of their normalized utilities and distances to previous offers
MultiplicativeLastOfferFollowingTBNegotiator(*args)	A time-based negotiator that selectes outcomes from the list allowed by the current utility level based on a weighted sum of their normalized utilities and distances to previous offers
AdditiveLastOfferFollowingTBNegotiator(*args)	A time-based negotiator that selectes outcomes from the list allowed by the current utility level based on a weighted sum of their normalized utilities and distances to previous offers
MultiplicativeFirstFollowingTBNegotiator(*args)	A time-based negotiator that selectes outcomes from the list allowed by the current utility level based on a weighted sum of their normalized utilities and distances to previous offers
AdditiveFirstFollowingTBNegotiator(*args[, ...])	A time-based negotiator that selectes outcomes from the list allowed by the current utility level based on a weighted sum of their normalized utilities and distances to previous offers
NaiveTitForTatNegotiator(*args[, kindness, ...])	Implements a naive tit-for-tat strategy that does not depend on the availability of an opponent model.
SimpleTitForTatNegotiator	alias of NaiveTitForTatNegotiator
RandomNegotiator([p_acceptance, ...])	A negotiation agent that responds randomly in a single negotiation.
RandomAlwaysAcceptingNegotiator([...])
NiceNegotiator(*args, **kwargs)	Offers and accepts anything.
CABNegotiator(*args, **kwargs)	Conceding Accepting Better Strategy (optimal, complete, but not an equilibirum)
CARNegotiator(*args, **kwargs)	Conceding Accepting Rational Strategy (neither complete nor an equilibrium)
CANNegotiator(*args, **kwargs)	Conceding Accepting Not Worse Strategy (optimal, complete, but not an equilibirum)
WABNegotiator(*args, **kwargs)	Wasting Accepting Better (neither complete nor an equilibrium)
WARNegotiator(*args, **kwargs)	Wasting Accepting Any (an equilibrium but not complete)
WANNegotiator(*args, **kwargs)	Wasting Accepting Any (an equilibrium but not complete)
MiCRONegotiator(*args[, accept_same])	Rational Concession Negotiator
EvaluationStrategy()
LocalEvaluationStrategy()
AnyAcceptEvaluationStrategy(strategies)
AllAcceptEvaluationStrategy(strategies)
GAOEvaluationStrategy()
TAUEvaluationStrategy([n_outcomes, ...])	Implements the Tentative-Accept Unique-Offers Generalized Bargaining Protocol.
OfferingConstraint()
LocalOfferingConstraint()
AnyOfferingConstraint(constraints)
AllOfferingConstraints(constaints)
RepeatFinalOfferOnly([n])
RepeatLastOfferOnly([n])
UniqueOffers()

Class Inheritance Diagram