English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Graph problems such as traveling salesman problem, or finding minimal Steinertrees are widely studied and used in data engineering and computer science.Typically, in real-world applications, the features of the graph tend to changeover time, thus, finding a solution to the problem becomes challenging. Thedynamic version of many graph problems are the key for a plethora of real-worldproblems in transportation, telecommunication, and social networks. In recentyears, using deep learning techniques to find heuristic solutions for NP-hardgraph combinatorial problems has gained much interest as these learnedheuristics can find near-optimal solutions efficiently. However, most of theexisting methods for learning heuristics focus on static graph problems. Thedynamic nature makes NP-hard graph problems much more challenging to learn, andthe existing methods fail to find reasonable solutions.  In this paper, we propose a novel architecture named Graph Temporal Attentionwith Reinforcement Learning (GTA-RL) to learn heuristic solutions forgraph-based dynamic combinatorial optimization problems. The GTA-RLarchitecture consists of an encoder capable of embedding temporal features of acombinatorial problem instance and a decoder capable of dynamically focusing onthe embedded features to find a solution to a given combinatorial probleminstance. We then extend our architecture to learn heuristics for the real-timeversion of combinatorial optimization problems where all input features of aproblem are not known a prior, but rather learned in real-time. Ourexperimental results against several state-of-the-art learning-based algorithmsand optimal solvers demonstrate that our approach outperforms thestate-of-the-art learning-based approaches in terms of effectiveness andoptimal solvers in terms of efficiency on dynamic and real-time graphcombinatorial optimization.

Solving Dynamic Graph Problems with Multi-Attention Deep Reinforcement Learning