Transfer Learning Based Adaptive Automated Negotiating Agent Framework
Authors: Ayan Sengupta, Shinji Nakadai, Yasser Mohammad
IJCAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Finally, we present the results of an agent generated using our framework in different ANAC domains with 88 different utility functions each and show that our agent outperforms the benchmark score by domain independent agents by 6%. Section 5 outlines the experimental setup, Section 6 shows the evaluations of our framework and finally, we conclude with Section 7 by discussing the limitations and provide directions for future work. |
| Researcher Affiliation | Collaboration | Ayan Sengupta1 , Shinji Nakadai1,2 , Yasser Mohammad1,2,3 1NEC Corporation, Japan 2National Institute of Advanced Industrial Science and Technology (AIST), Japan 3Assiut University, Egypt |
| Pseudocode | Yes | Algorithm 1 Single metric based Critic algorithm for opponent utility function |
| Open Source Code | No | The paper does not provide an explicit statement or a link to the open-source code for the methodology developed in this paper. It mentions using 'Tensorflow' and 'NegMAS library', which are third-party tools, but not their own implementation. |
| Open Datasets | Yes | We analyzed our proposed framework using 3 ANAC 2015 domains with 88 opponent utility functions and 88 self utility functions for each domain. |
| Dataset Splits | No | The paper discusses training of the models but does not provide specific details on how the dataset was split into training, validation, and test sets, such as percentages or sample counts. It only refers to 'training the Base Model' and 'Hsmall for training'. |
| Hardware Specification | No | The paper does not provide specific hardware details such as GPU/CPU models, processor types, or memory used for running the experiments. |
| Software Dependencies | No | The paper mentions using 'Tensorflow [Abadi et al., 2015]' and 'Neg MAS [Mohammad et al., 2021]' libraries for deep learning models and simulations, respectively. However, it does not provide specific version numbers for these software dependencies or other key components. |
| Experiment Setup | Yes | In this section we briefly describe the experimental setup for our evaluations. ... During training we used weighted cross-entropy loss function with κ = 4. ... for our experiments in section 5 we used XGBoost[Chen and Guestrin, 2016] based classifier for training the Critic... Our experiments also include another version of Critic where we use Algorithm 1 with Wasserstein Distance and αM as 3, 82 and 538 for the domains Bank Robbery, car domain and Tram respectively. ... Next, for the condition when self utility functions are changing we used the Algorithm 2 with γ1 = 0.7, γ2 = 1 and αL = 0.2. |