End-to-End Goal-Driven Web Navigation

By Rodrigo Nogueira, Kyunghyun Cho. 2016

41 citations recorded by the source corpus.

We propose a goal-driven web navigation as a benchmark task for evaluating an agent with abilities to understand natural language and plan on partially observed environments. In this challenging task, an agent navigates through a website, which is represented as a graph consisting of web pages as nodes and hyperlinks as directed edges, to find a web page in which a query appears. The agent is required to have sophisticated high-level reasoning based on natural languages and efficient sequential decision-making capability to succeed. We release a software tool, called WebNav, that automatically transforms a website into this goal-driven web navigation task, and as an example, we make WikiNav, a dataset constructed from the English Wikipedia. We extensively evaluate different variants of neural net based artificial agents on WikiNav and observe that the proposed goal-driven web navigation well reflects the advances in models, making it a suitable benchmark for evaluating future progress. Furthermore, we extend the WikiNav with question-answer pairs from Jeopardy! and test the proposed agent based on recurrent neural networks against strong inverted index based search engines. The artificial agents trained on WikiNav outperforms the engined based approaches, demonstrating the capability of the proposed goal-driven navigation as a good proxy for measuring the progress in real-world tasks such as focused crawling and question-answering.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Natural Language Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Natural Language Processing

Citations in the corpus

WebShop: Towards Scalable Real-World Web Interaction with Grounded Language Agents (2022, background): “Nogueira and Cho [36] introduced WikiNav as a benchmark for RL agents navigating pages, but the task is purely navigational with the actions restricted to either choosing a hyperlink to follow or deciding to stop.”
The Web as a Knowledge-Base for Answering Complex Questions (2018, methodology): “Paraphrasing for better interaction with a QA model has been recently suggested by Buck et al. (2017) and Nogueira and Cho (2016).”
Ask the Right Questions: Active Question Reformulation with Reinforcement Learning (2017, background): “The work of Nogueira and Cho [2016] is most related to ours.”
Deep Reinforcement Learning with a Natural Language Action Space (2015, background): “More recently, Nogueira and Cho (2016) have also proposed a goal-driven web navigation task for language based sequential decision making study.”
Understanding HTML with Large Language Models (2022, background): “Web crawling (Olston et al., 2010), form-ﬁlling (Diaz et al., 2013; Gur et al., 2021), or information retrieving web agents (Nogueira and Cho, 2016) are important for both automating and assisting users in web-based tasks.”
Deep Reinforcement Learning: An Overview (2017, background): “VIN can generalize in a diverse set of tasks: si mple gridworlds, Mars Rover Navigation, continuous control and WebNav Challenge for Wikipedi a links navigation (Nogueira and Cho, 2016).”
Learning to Coordinate Multiple Reinforcement Learning Agents for Diverse Query Reformulation (2018, methodology): “J EOPARDY : This dataset was introduced by Nogueira & Cho (2016).”
Multi-agent query reformulation: Challenges and the role of diversity (2019, methodology): “Jeopardy: This dataset was introduced by Nogueira & Cho (2016).”
Deep Graph Library: Towards Efficient and Scalable Deep Learning on Graphs (2019, background): “…traversal order (Tai et al., 2015; Jin et al., 2018); • random walk (Ying et al., 2018a; Hamilton et al., 2017); • active sets that are generated dynamically, including exploration algorithms on graph navigation (Nogueira & Cho, 2016; Das et al., 2017) and graph generative models (Li et al., 2018).”
Language Models can Solve Computer Tasks (2023, unknown): “Although the results presented in this paper are only on a research benchmark, if we extrapolate forward the capabilities of these models and methods, we anticipate vast broader impacts that have the potential to revolutionize numerous industries.”