IARAI (The Institute for Advanced Research in Artificial Intelligence), which is an independent global machine learning research institute established by HERE Technologies, announced the results and winners of its traffic prediction competition, aimed to solve mobility challenges using artificial intelligence (AI).
“Traffic4cast” is a unique competition merging movie prediction machine learning with traffic research, which challenged the competitors to understand the complex traffic systems and make the predictions about how they would flow in the future.
Results reveals that how AI can effectively uncover the insights to solve the traffic gridlock through trial and error of industrial geospatial data from HERE, which is a leader in mapping and location-based services.
The Traffic comes about when the drivers make simple decisions lead to complex behavior patterns. And these patterns depend on various factors such as time of day, and the road network, the congestion situations, holidays, weather conditions and day of the week. In order to effectively identifying and analyzing the traffic patterns lead to more accurate predictions of how traffic would move on the given roads at given times of day.
AI (Artificial Intelligence) and more specifically the neural networks, computer systems modeled on human brain and nervous system can help to solve this problem, because they are very good at spotting patterns. The neural networks “learn” to do tasks by considering the examples, such as datasets, and usually without being programmed with the task specific rules. That ability to learn without being programmed, means that although the neural networks are good at identifying patterns, why they are good at it is still unclear. Their inner workings are still one of the mysteries of machine learning, the so-called “black box” AI, meaning that the processes cannot be easily understood or tested by programmers.
The Traffic4cast results show that neural networks were the most effective method used at predicting traffic and came closest to simulating the exact traffic flow. All the top entrants used neural networks instead of “non-black box” solutions, such as support vector machines, Bayesian networks and other fixed algorithms. Winners from South Korea, Oxford/Zurich and Toronto were among more than 40 teams from around the world who submitted over 4,000 entries.