Automated negotiation is a very challenging research field that is gaining momentum in the e-business domain. There are three main categories of automated negotiations, classified according to the participating agent cardinality and the nature of their interaction (Jennings, Faratin, Lomuscio, Parsons, Sierra, & Wooldridge, 2001): the bilateral, where each agent negotiates with a single opponent, the multi-lateral which involves many providers and clients in an auction-like framework and the argumentation/persuasion-based models where the involving parties use more sophisticated arguments to establish an agreement. In all these automated negotiation domains, several research efforts have focused on predicting the behaviour of negotiating agents. This work can be classified in two main categories. The first is based on techniques that require strong a-priori knowledge concerning the behaviour of the opponent agent in previous negotiation threads. The second uses mechanisms that perform well in single-instance negotiations, where no historical data about the past negotiating behaviour of the opponent agent is available. One quite popular tool that can support the latter case is Neural Networks (NNs) (Haykin, 1999). NNs are often used in various real world applications where the estimation or modelling of a function or system is required. In the automated negotiations domain, their usage aims mainly to enhance the performance of negotiating agents in predicting their opponents’ behaviour and thus, achieve better overall results on their behalf. This paper provides a survey of the most popular automated negotiation approaches that are using NNs to estimate elements of the opponent’s behaviour. The rest of this paper is structure as follows. The second section elaborates on the state of the art bilateral negotiation frameworks that are based on NNs. The third section briefly presents the multilateral negotiation solutions that exploit NNs. Finally, in the last section a brief discussion on the survey is provided.
Neural Networks In Bilateral Negotiations
In (Zhang, Ye, Makedon, & Ford, 2004) a hybrid bilateral negotiation strategy mechanism is described that supplies negotiation agents with more flexibility and robustness in an automated negotiation system. The framework supports a dynamically assignment of an appropriate negotiation strategy to an agent according to the current environment, along with a mechanism to create new negotiation rules by learning from past negotiations. These learning capabilities are based on feedforward back-propagation neural networks and multidimensional inter-transaction association rules. However, the framework is not adequately described and defined and the neural networks are not specifically instantiated. Additionally, there are neither quantitative nor qualitative experimental results for real world cases. Finally, the format of the input to the generic network that is presented is ambiguously described.
In (Zeng, Meng, & Zeng, 2005), the authors employ a neural network to assist the negotiation over very specific issues from a real world example. The network is trained online by the past offers made by the opponent, while both the buyer and the seller agent have the ability to employ the proposed network. However, the experimental data sets are very restrictive and do not address the diversity of those that can be arisen in real scenarios. Additionally, the authors do not present the actual size of the hidden layer, a parameter that is extremely crucial with regards to the appropriateness to use such a network in a real time negotiation procedure by an agent with limited resources.
Key Terms in this Chapter
Bilateral Negotiation: A negotiation procedure, where exactly two parties are involved, i.e. a client and a provider.
Multi-Layer Perceptron (MLP): A fully connected feedforward NN with at least one hidden layer that is trained using back-propagation algorithmic techniques.
Neural Network (NN): A network modelled after the neurons in a biological nervous system with multiple synapses and layers. It is designed as an interconnected system of processing elements organized in a layered parallel architecture. These elements are called neurons and have a limited number of inputs and outputs. NNs can be trained to find nonlinear relationships in data, enabling specific input sets to lead to given target outputs.
Multilateral Negotiation: A negotiation procedure, where more than two parties are involved, i.e. multiple clients and/or providers negotiate simultaneously.
RBF NN: It is an artificial NN, the activation functions of which are radial basis functions. It has two layers of processing, where the first maps the input onto each RBF neuron in the other (hidden) layer.
Automated Negotiation: It is the process by which group of actors communicate with one another aiming to reach to a mutually acceptable agreement on some matter, where at least one of the actors is an autonomous software agent.
Radial Basis Function (RBF): Function that involves a distance criterion with respect to a centre, such as a circle, ellipse or Gaussian.