You are on page 1of 35

Negotiating Agents

Vishay Raina
M.Tech, SE
• Negotiation among autonomous
computational agents: principles, analysis and
challenges. (2008)
• When Will Negotiation Agents Be Able to
Represent Us? The Challenges and
Opportunities for Autonomous Negotiators
Negotiation among autonomous computational agents:
principles, analysis and challenges
Fernando Lopes · Michael Wooldridge · A. Q. Novais
• Introduction:
• Negotiation is an important form of social interaction,
businesses negotiate to purchase raw materials and to sell
• Demands for systems composed of computational agents
owned by different individuals or organizations and are
capable of reaching agreements through negotiation are
becoming increasingly important and pervasive.
• Examples:
1. the industrial trend toward agent-based supply chain
2. the business trend toward virtual enterprises
2.1 Negotiation in artificial intelligence
• Two main perspectives: a theoretical or formal mathematical
perspective and a practical or system-building perspective
• The theoretical perspective draws heavily from game-
theoretic and economic methods.
• highly desirable properties, such as Pareto efficiency and the
ability to guarantee convergence,
• abstract problems under assumptions that often limit their
• primarily focused on formal bargaining, auctions, market-
oriented programming, contracting, and coalition formation.
• the practical perspective draws heavily on social sciences
techniques for understanding interaction and negotiation.
• primarily focused on the central process of moving toward
• Most theoretical models make the restrictive assumptions
which fail in most realistic environments due to the limited
processing and communication capabilities of existing
• Most computational models are being used successfully in a
wide variety of real-world domains.
• Two fundamental approaches to the study of human
negotiation: the game-theoretic approach and the
behavioural approach . These two approaches have motivated
the distinction between theoretical and computational
• The authors discuss and analyze three classes of models:
1. Game-theoretic models:
– provide clear analysis of specific negotiation situations
and precise results concerning the optimal strategy
– i.e., the strategy that maximizes negotiation outcome.
– Negotiation proceeds by an iterative exchange of
proposals and counterproposals
2. Heuristic models:
– provide general guidelines to assist negotiators and
beneficial strategies for moving toward agreement
– i.e., strategies that lead to good (rather than optimal)
3. Argumentation-based models:
– allow negotiators to argue about their mental attitudes
during the negotiation process
– negotiators can provide arguments to: (1) justify their
negotiation stance, or (2) persuade other negotiators to
change their negotiation stance
– negotiation may proceed by an iterative exchange of
proposals, counterproposals, threats, promises, etc.
2.2 Agents for supply chain
• Multi-agent systems (MAS) are ideally suited for
problems with multiple problem solving entities and
multiple problem solving methods
• Central to the design and effective operation of a
multi-agent system are a core set of problems and
research questions:
1. the design problem: how to formulate, describe,
different problems and synthesize results among a group
of intelligent agents?
2. the coordination problem: how to ensure that agents act
coherently, accommodating the local decisions or non-
local effects and avoiding harmful interactions?
Supply chain
• A supply chain is a network of facilities that performs the
– procurement of raw materials from suppliers,
– transformation into intermediate goods and final products
– delivery of these products to customers.
• A multi-agent supply chain system is
– a collection of autonomous computational agents,
– each responsible for one or more supply chain functions,
– interacting with other agents in the execution of their
• Supply chain functions range from the ordering and receipt of
raw materials to the distribution and delivery of final
products, via the scheduling, production, and warehousing of
intermediate goods and final products
A typical distribution involves at least the
following agents:
1. sales agent: responsible for acquiring orders from
customers, negotiating with customers, and handling
customer requests for modifying or cancelling orders.
2. logistics agent: coordinating the plants and distribution
centers of a manufacturing enterprise. It manages the
movement of materials and products across the supply
chain, from the suppliers of raw materials to the customers
of finished goods.
3. scheduling agent: responsible for scheduling and
rescheduling the activities of a manufacturing enterprise.
4. resource management agent: responsible for dynamically
managing the availability of resources in order to execute
the scheduled activities.
5. supplier agents and customer agents: the suppliers sell raw
materials and the customers buy finished goods.
• The opening stance and the pattern of
concessions are two central elements of
• Negotiators who demand too much will often
fail to reach agreement and thereby do poorly.
• Those who demand too little will usually reach
agreement but achieve low benefits
A specific situation involving negotiation
between the sales agent and the logistics agent
• David, the director of Sales, has lined up
– two new orders for a total of 15,000 men’s suits: one for
10,000 and the other for 5,000 men’s suits.
• Martin, the director of Logistics, has already stated that it will
take four months to make the suits
• The problem is that Martin insists that the job will take four
months and David’s customer wants a 2-month turnaround.
• They can resolve their differences by negotiating a mutually
beneficial agreement.
• If David is overly firm and keeps on demanding a two-month
schedule, negotiation may break down. But if he is overly soft
and concedes a lot, they will lose the orders.
• Four major issues of concern: quantity_1, date_1, quantity_2
and date_2.
• Depending on the priorities, decisions can be made.
• It is of higher priority for Sales to get fast action on the 10,000
suit order than the 5000 suit order.
• Suppose now that it is of higher priority for Logistics to avoid
handling the 10,000 suit order.
– Logistics prefers handling the minimal number of suits that
Sales is willing to accept (4,000 suits in 6 weeks).
– logrolling deal—each party can yield on issues that are of
low priority to itself and high priority to the other side.
• Accordingly, they can agree on the following superior
solution: a 4-week schedule for 10,000 suits and a 6-week
schedule for 4,000 suits.
A generic framework for automated negotiation

• Negotiation is a discussion among conflicting parties

with the aim of reaching agreement about a
divergence of interest
• list of situations:
• purely competitive: completely opposed interests
• purely cooperative, as when the parties have perfectly
compatible interests
• mixed-motive, containing elements of both
competitive and cooperative situation
• Conflict, is present in most negotiation situations, it is
the basis or the driving force of negotiation
Distinct phases: sequences of events that
constitute the story of negotiation
• Three phases:
– a beginning or initiation phase,
– middle or problem-solving phase,
– an ending or resolution phase
• Initiation phase focuses on the preparation and planning for
– each party’s efforts to emphasize points of difference and
posture for positions.
• Problem solving phase seeks a solution for a dispute
– extensive interpersonal interaction, movement toward a
mutually acceptable agreement.
• resolution phase focuses on details and implementation of a
final agreement.
dimensions of a generic framework for
automated negotiation:
Group dimension
Preliminaries 1. Social conflict (detection and
2. Negotiating parties (number of
Pre-negotiation 3. Structuring of personal
information (definition and
execution of key pre-negotiation
4. Analysis of the opponents
(gathering and use of key
5. Definition of the protocol and
selection of the initial strategy
Group Dimension
Actual negotiation 6. Exchange of offers and
feedback information
7. Argumentation (exchange of
threats, promises, etc.)
8. Learning (in negotiation)
9. Dynamic strategic choice
(selection of new strategies)
10. Impasse resolution
Renegotiation 11. Analysis and improvement
of the final agreement.
When Will Negotiation Agents Be Able to
Represent Us? The Challenges and
Opportunities for Autonomous Negotiators
• Introduction
– benefits that computerized negotiation can offer,
– better (win-win) deals, and reduction in time, costs, stress
and cognitive effort on the part of the user.
• Example:
– the smart electrical grid, flexible devices in our household
will soon (re-)negotiate complex energy contracts
– Another example is the rise of the Internet of Things (IoT),
which will introduce countless smart, interconnected
devices that autonomously negotiate the usage of
sensitive data and make trade-offs between privacy
concerns, price, and convenience.
• While many successes have been achieved truly
autonomous negotiators are still a thing of the
• This begs the obvious question: what is still
lacking currently and what is needed for
autonomous negotiators to be able to fulfill their
• This paper discusses the challenges and
upcoming application domains for (almost)
entirely autonomous negotiation on people’s
The Autonomy Diagonal of Negotiation
Negotiation research clustered
around one of the three main
orthogonal dimensions of
• Self-sufficiency: the capability
of the actor to take care of
• Self-directedness: the
freedom to act within the
environment and the means to
reach goals.
• Support for interdependence:
being able to work with others
and influence and be
influenced by team members
The Autonomy Diagonal of Negotiation

Three strands of research in

automated negotiation
•Negotiation support systems:
These systems are designed to
assist and train people in
•Enable interdependence by
design, humans predominately
supervise and make decisions
on the appropriate outcome,
which results in low self-
sufficiency and self-
The Autonomy Diagonal of Negotiation
Game theoretical approaches:
•Game theory’s dominant concern is
with fully rational players and what
each should optimally do.
•The focus is equilibrium strategies
or protocols that can guarantee win-
•Examples :
•high frequency trading agents for
financial exchanges, advertising
exchanges, or sniping agents used in
•function without human
intervention, highly self-sufficient,
constrained in terms of freedom to
direct the process.
The Autonomy Diagonal of Negotiation
Negotiation analytical
•Negotiation analysis prescribes
how players should act given a
description of how others will
•A key feature: to construct
beliefs based on partial
information and act in best
response to this belief, typically
over opponent types or
•This locates the negotiation
analytical approach around the
self-directed axis.
• Argument that the main automated
negotiation research lines have developed in
parallel to one of the three autonomy
• Sound strategy
• We can make substantive progress in
autonomous negotiation by continuing to
advance along the autonomy diagonal
3 Major Challenges
Major Main Building Exampl Solutions roadmap
challenge autono blocks e
my opport
dimensi unities
Domain Self- Preference Privacy Value of information
knowledge sufficie elicitation and IoT indicators, robust
and ncy & on-the-fly performance
preference Interde estimates
elicitation penden
Domain Smart Separate user/agent
modeling grids domain models, expert
Long-term Self- Repeated Communiti Temporally
perspectiv sufficie interactio es, smart integrative
e ncy & ns homes, negotiations,
Self- autonomo reputation metrics
directe us driving
Non- B2B, Cost-efficient
stationary entertainm tracking, context
preferenc ent dependent
es booking models,
User Self- Acceptabi Conflict Co-creation,
trust and directedn lity and resolution, adjustable
adoption ess & participat customer autonomy, transfer
Interdepe ion retainment of control
Transpare Sharing Transparency and
nt economy, openness,
conseque decentraliz
nces ed
Preference elicitation on-the-fly
• an autonomous negotiator needs to engage with the user
to make sure it constructs an accurate preference model
• domains where people reluctant to engage with the
system, for instance in privacy negotiations.
• automated negotiators of the future are required to not
only strike deals with limited available user information,
but also to assess which additional information should be
elicited from the user, while minimizing user bother
• Belief that future research should particularly emphasize
preference elicitation on-the-fly: that is, active preference
extraction during negotiation(s).
Long term perspective
• Repeated encounters:
• Example: community energy agents can trade energy from
storage and local sources between neighbouring homes and
• Another example is the smart home, where different
occupants will have different needs and preferences and have
to reach mutual agreements, e.g. about the temperature of
the house and the use of devices
• self-driving vehicles, where vehicle-to-vehicle and vehicle to-
infrastructure negotiation can play an important role (e.g.,
negotiating priority at intersections).
• Problem: Negotiation opportunities for isolated encounters
can be very limited, since often a resource (e.g. electricity or
giving way) is needed without necessarily offering anything
immediately in return (except possibly money or virtual
• Explicitly considering the temporal dimension allows agents to
receive or concede something now in return for conceding or
receiving the same resource later. Single-issue, distributive
negotiations can be turned into richer, multi-issue, integrative
negotiations, with more scope to achieve win-win solutions
• significant challenge: future needs are often uncertain, and so
it is difficult to commit to giving up or requesting specific
future resources.
• Possible solutions involve money or virtual currencies which
can be redeemed at a later stage.
• Another possible solution is to rely on altruism and using trust
ratings and reputation systems to provide the desired
• For example, in the context of organ donation, t
– he equity principle would allocate resources on the basis
of ability, effort or merit,
– the equality rule would treat individuals the same
– principle of need is usually achieved by allocating
according to individuals medical condition,
• One approach is to incorporate psychological factors into the
utility function
Non-stationary preferences
• In long-term negotiations, these very preferences
may evolve over the course of weeks or months
according to certain preference dynamics.
• It will erroneously fulfill outdated design
• This reframes the challenge posed in “Preference
elicitation on-the-fly” of preference elicitation to
“cost-efficient tracking of non-stationary
preferences” in long-term negotiation
User Trust and Adoption
• While the agent depends on the user for knowledge and
guidance the user relies on a self-directed agent for a good
• User participation:
• each negotiation party consists of at least one human and
one negotiation agent.
• The agent should do the brunt of the negotiation work to
find possible agreements with the other negotiation parties
and which can presented to their human partners for
feedback and new input.
• The research challenge is to determine when, how, and
how often to switch the initiative from human to agent and
vice versa.
User Trust and Adoption
• Transparent consequences : There is an inherent tension between
increased self-directedness and trust, which dampens the adoption
of increasingly autonomous negotiators: on the one hand, an
autonomous negotiatior’s relevance is directly proportional to its
ability to impact the user independently in meaningful ways
• the user’s trust and willingness to relinquish control is conditional
on understanding the agent’s reasoning and consequences of its
• The two can be reconciled by making the outcome space more
transparent to the user, and by enabling the user to specify the
permissible means in the form of principles.
• They argue unpredictable behavior is in fact desirable as a
negotiation tactic as a confusing and randomization device, as long
as the consequences are transparently explained to the user
Thank you