Rationality and Self-Interest in Peer
to Peer Networks
By: Jeffrey Shneidman and David C.
Parkes
Paper Abstract:
Much of the existing work in peer to
peer networking assumes that users will follow prescribed protocols without
deviation. This assumption ignores the user’s ability to modify the behavior of
an algorithm for self-interested reasons.
We advocate a different model in
which peer to peer users are expected to be rational and self- interested. This
model is found in the emergent fields of Algorithmic Mechanism Design (AMD) and
Distributed Algorithmic Mechanism Design (DAMD), both of which introduce
game-theoretic ideas into a computational system. We, as designers, must create
systems (peer to peer search, routing, distributed auctions, resource
allocation, etc.) that allow nodes to behave rationally while still achieving
good overall system outcomes.
This paper has three goals. The first
is to convince the reader that rationality is a real issue in peer to peer
networks. The second is to introduce mechanism design as a tool that can be
used when designing networks with rational nodes. The third is to describe
three open problems that are relevant in the peer to peer setting but are
unsolved in existing AMD/DAMD work. In particular, we consider problems that
arise when a networking infrastructure contains rational agents.
Learners Notes and Synthesis:
In our social environment every human
being tends to protect its own self interest and survival at all cause. In
order for a society to grow we should learn to be govern by protocols and
rules. So with peer-to-peer network we cannot assume that every nodes will
follow desired protocols establish by the central authority or the
developer of the application.
Some peer to peer application are
govern by rules and protocols:
* Routing to reach other peer
* Resource management
* Decision making based on facts and
environment behavior
* Distribute load among other peer
Rationality vs Self interest
Based on the example on this paper,
running the auction on a large peer-to-peer network, initially you will. Be
announcing the bidding for the auction and you are waiting for a lot of bids.
But unfortunately there are only three bidders only to find out that they are
your direct neighbors, for their self interest of this neighbor they did
not forward the advertisement for the auction. This example illustrates
basic problem of peer-to-peer network for rationality vs self
interest.
Rationality in peer-to-peer network
Free rider problem:
“A free rider receives the benefit of everyone else's
cooperation without having to cooperate himself. Think of a single person in
the community who doesn't pay his taxes; he gets all the benefits of the public
institutions those taxes pay for—police and fire departments, road construction
and maintenance, regulations to keep his food and workplace safe, a
military—without having to actually pay for them.” [1]
Tragedy of the commons problems:
“A Tragedy of the Commons occurs whenever a group shares a
limited resource: not just fisheries, but grazing lands, water rights, time on
a piece of shared exercise equipment at a gym, an unguarded plate of cookies in
the kitchen. In a forest, you can cut everything down for maximum short-term
profit, or selectively harvest for sustainability. Someone who owns the forest
can make the trade-off for himself, but when an unorganized group together owns
the forest there's no one to limit the harvest, and a Tragedy of the Commons
can result” [1]
In peer-to-peer system nodes that do
not produce the same level of their consumption are considered leechers, while
peers that share the their resources to provide a better performance.
Another example peer to peer clients
that deflect from protocols for their own interest are the user that develop
their own client program that will deflect on the protocols and will circumvent
the network for their own self interest.
Rational users can free ride on
systems, create software designed to subvert mechanisms, and generally place
themselves into selfishly advantageous situations.
Reason for this peer-to-peer network
still works is that are enough number of peers that obediently
follow the prescribes rules and protocols. Example are users that
download obedient program that will be used for the network.
Mechanism design:
Algorithmic Mechanism Design (AMD)
addresses the computational complexity of the mechanism infrastructure and
attempts to construct mechanisms that produce desired outcomes while retaining
computational feasibility.
Distributed Algorithmic Mechanism
Design (DAMD) is an even newer construction
Nodes type in a peer-to-peer are the
following:
1. Correct nodes this are the nodes
that follow prescribe protocols by the system
2. Faulty nodes have been classified that
stops from working either stop from sending and receiving node that are
dropping messages
3. Rational nodes are
classified as those nodes that maximized their utility in a given
environment, they tend to learn from their neighbors then use this for their
benefit.
4. Irrational nodes behave rationally
but dot not follow the mechanism design by the developer. They are node that
behave irrationally considering only their own preferences sometimes they are
called the antisocial nodes.
Conclusion:
Rationality is a concern on
peer-to-peer system.
Without deflectance a system will
become stagnant and no improvement. Challenge is how to we ensure
compliance and reward to the nodes that follow the protocol in the
peer-to-peer system which in-turn provide better performance for the whole
system.
References:
1. Excerpt From: Schneier, Bruce.
“Liars and Outliers: Enabling the Trust That Society Needs to Thrive.” Wiley,
2012-02-14T05:00:00+00:00.
2. Eytan Adar and Bernardo Huberman.
Free Riding on Gnutella. First Monday, 5(10), October 2000.
3. Garrett Hardin. The Tragedy of the
Commons. Science, 162:1243–1248, 1968. Alternate Location: http:// dieoff.com/page95.htm.
No comments:
Post a Comment