Building Personalized Simulator for Interactive Search

Authors: Qianlong Liu, Baoliang Cui, Zhongyu Wei, Baolin Peng, Haikuan Huang, Hongbo Deng, Jianye Hao, Xuanjing Huang, Kam-Fai Wong

IJCAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results based on real-world dataset demonstrate the effectiveness of our agent and personalized simulator. and 5 Experiments 5.1 Dataset and Abstract Query 5.2 Experiment Setup 5.3 Experimental Results
Researcher Affiliation Collaboration 1Fudan University, China 2Alibaba Group, China 3The Chinese University of Hong Kong, Hong Kong 4Tianjin University, China
Pseudocode Yes Algorithm 1: Building real experience buffer with abstract query., Algorithm 2: Rollout Simulation, and Algorithm 3: Training Algorithm
Open Source Code No The paper does not provide any explicit statement or link indicating the open-sourcing of the code for the described methodology.
Open Datasets No Our dataset is derived from the log of Taobao APP 2, which is processed into transition tuple, i.e., (s, a, r, s , trmt). 2The largest E-commerce platform in China.
Dataset Splits Yes Table 1: Description of our dataset, of which 60% for training, 20% for testing, 20% for validation.
Hardware Specification No The paper does not provide specific details about the hardware used to run the experiments, such as GPU models, CPU types, or memory specifications.
Software Dependencies No The paper mentions 'word2vec' but does not specify its version or any other software dependencies with version numbers.
Experiment Setup Yes Implementation Details. The simulator is pre-trained before interacting with our agent and fixed unchanged while interacting with the agent. Each word is represented by word embedding (200-dim) trained on the historical queries from other scenarios of Taobao APP ( 12.7 million queries, 332,922 words) via word2vec. ϵ = 0.2, γ = 0.9, m = 5, N s = 5, T = 20, learning rate is set to 10 5 and 10 3 for the training of environment simulator and agent respectively. At each turn, the agent recommends 3 tags to the user (i.e., K = 3). The target network of the agent is updated every 200 steps. The hidden size (both bidirectional LSTM and fully connected layers) of simulator and agent is 5 and 10. The length of simulated experiences buffer Ds is 1000.