Back to Table of contents

Primeur weekly 2016-08-22

Special

ExaCT team shows how Legion S3D code is a tribute to co-design on the way to exascale supercomputing ...

Focus

Sunway TaihuLight's strengths and weaknesses highlighted by Jack Dongarra ...

Exascale supercomputing

Big PanDA tackles Big Data for physics and other future extreme scale scientific applications ...

Computer programming made easier ...

Quantum computing

Cryptographers from the Netherlands win 2016 Internet Defense Prize ...

Focus on Europe

STFC Daresbury Laboratory to host 2016 Hands-on Tutorial on CFD using open-source software Code_Saturne ...

Middleware

Germany joins ELIXIR ...

Columbus Collaboratory announces CognizeR, an Open Source R extension that accelerates data scientists' access to IBM Watson ...

Cycle Computing optimizes NASA tree count and climate impact research ...

GPU-accelerated computing made better with NVIDIA DCGM and PBS Professional ...

Hardware

Mellanox demonstrates accelerated NVMe over Fabrics at Intel Developers Forum ...

Nor-Tech has developed the first affordable supercomputers designed to be used in an office, rather than a data centre ...

NVIDIA CEO delivers world's first AI supercomputer in a box to OpenAI ...

AMD demonstrates breakthrough performance of next-generation Zen processor core ...

CAST and PLDA Group demonstrate x86-compliant high compression ratio GZIP acceleration on FPGA, accessible to non-FPGA experts using the QuickPlay software defined FPGA development tool ...

IBM Research - Almaden celebrates 30 years of innovation in Silicon Valley ...

Wiring reconfiguration saves millions for Trinity supercomputer ...

Cavium completes acquisition of QLogic ...

Applications

Soybean science blooms with supercomputers ...

NOAA launches America's first national water forecast model ...

Computers trounce pathologists in predicting lung cancer type, severity, researchers find ...

Star and planetary scientists get millions of hours on EU supercomputers ...

Bill Gropp named acting director of NCSA ...

Latest NERSC/Intel/Cray dungeon session yields impressive code speed-ups ...

User-friendly language for programming efficient simulations ...

New book presents how deep learning neural networks are designed ...

Liquid light switch could enable more powerful electronics ...

Energy Department to invest $16 million in computer design of materials ...

Pitt engineers receive grant to develop fast computational modelling for 3D printing ...

Environmental datasets help researchers double the number of microbial phyla known to be infected by viruses ...

Teaching machines to direct traffic through deep reinforcement learning ...

Simulations by PPPL physicists suggest that magnetic fields can calm plasma instabilities ...

New material discovery allows study of elusive Weyl fermion ...

New maths to predict dangerous hospital epidemics ...

Kx financial analytics technology tackles Big Data crop research at biotech leader Earlham Institute ...

The Cloud

New hacking technique imperceptibly changes memory virtual servers ...

Teaching machines to direct traffic through deep reinforcement learning

16 Aug 2016 Beijing - Rush hour. The dreaded time of day when traffic conditions seem bent on making you late. As your car slowly creeps in line behind countless others stuck at a stop light, you think to yourself, "Why aren't these lights changing faster?" Traffic control scientists have long tried to solve this signaling problem. Unfortunately, the complexity of traffic situations has made the job extremely hard. A recent study suggests that machines can learn how to plan traffic signals just right to reduce wait times and make traffic queues shorter.

Automating traffic control is notoriously tricky because it involves two challenging tasks: modelling traffic flow and then optimizing it. While traditional artificial intelligence and simulation techniques can learn and reproduce the dynamics of certain traffic situations, they cannot easily determine the best course of action. Despite this shortcoming, one emerging AI method that has shown promise is reinforcement learning.

Much like the reward system in our brain, reinforcement learning algorithms operate by determining what set of actions are most beneficial to a system in a given state. In video games, a popular proving ground for this type of algorithm, these steps are the strategic moves a player must make to earn the highest possible score. In traffic signaling, it's the pattern of traffic light switches that minimizes the time drivers spend waiting in a queue and/or the length of that queue. The problem with this algorithm is that the dataset it must comb through becomes exponentially larger as traffic conditions and ways of controlling traffic become increasingly complex. This is where another type of AI method called deep learning comes into play.

Also inspired by how the human brain works, deep learning uses special algorithms called neural networks to search for hidden patterns in sets of data. When combined with reinforcement learning, this technique effectively reduces the space a machine must search through as it seeks the best solution. This greatly reduces the required computational time.

When deployed on a virtual four-way traffic intersection with four lanes heading east-west and four heading north-south, the deep reinforcement learning algorithm outperformed a conventional reinforcement learning algorithm. There were shorter lines. There was a better balance of traffic in both directions. And over the course of a full day, more than 1000 fewer vehicles came to a full stop. In the end, the average delay was cut by 14%, with vehicles spending 13 seconds less in traffic during peak morning hours.

Deep reinforcement learning algorithms aren't quite ready to hit the streets. But they are helping make real-world traffic scenarios a lot less complicated for researchers, which could pave the way toward a new understanding of how traffic flows work.

Source: Chinese Association of Automation

Back to Table of contents

Primeur weekly 2016-08-22

Special

ExaCT team shows how Legion S3D code is a tribute to co-design on the way to exascale supercomputing ...

Focus

Sunway TaihuLight's strengths and weaknesses highlighted by Jack Dongarra ...

Exascale supercomputing

Big PanDA tackles Big Data for physics and other future extreme scale scientific applications ...

Computer programming made easier ...

Quantum computing

Cryptographers from the Netherlands win 2016 Internet Defense Prize ...

Focus on Europe

STFC Daresbury Laboratory to host 2016 Hands-on Tutorial on CFD using open-source software Code_Saturne ...

Middleware

Germany joins ELIXIR ...

Columbus Collaboratory announces CognizeR, an Open Source R extension that accelerates data scientists' access to IBM Watson ...

Cycle Computing optimizes NASA tree count and climate impact research ...

GPU-accelerated computing made better with NVIDIA DCGM and PBS Professional ...

Hardware

Mellanox demonstrates accelerated NVMe over Fabrics at Intel Developers Forum ...

Nor-Tech has developed the first affordable supercomputers designed to be used in an office, rather than a data centre ...

NVIDIA CEO delivers world's first AI supercomputer in a box to OpenAI ...

AMD demonstrates breakthrough performance of next-generation Zen processor core ...

CAST and PLDA Group demonstrate x86-compliant high compression ratio GZIP acceleration on FPGA, accessible to non-FPGA experts using the QuickPlay software defined FPGA development tool ...

IBM Research - Almaden celebrates 30 years of innovation in Silicon Valley ...

Wiring reconfiguration saves millions for Trinity supercomputer ...

Cavium completes acquisition of QLogic ...

Applications

Soybean science blooms with supercomputers ...

NOAA launches America's first national water forecast model ...

Computers trounce pathologists in predicting lung cancer type, severity, researchers find ...

Star and planetary scientists get millions of hours on EU supercomputers ...

Bill Gropp named acting director of NCSA ...

Latest NERSC/Intel/Cray dungeon session yields impressive code speed-ups ...

User-friendly language for programming efficient simulations ...

New book presents how deep learning neural networks are designed ...

Liquid light switch could enable more powerful electronics ...

Energy Department to invest $16 million in computer design of materials ...

Pitt engineers receive grant to develop fast computational modelling for 3D printing ...

Environmental datasets help researchers double the number of microbial phyla known to be infected by viruses ...

Teaching machines to direct traffic through deep reinforcement learning ...

Simulations by PPPL physicists suggest that magnetic fields can calm plasma instabilities ...

New material discovery allows study of elusive Weyl fermion ...

New maths to predict dangerous hospital epidemics ...

Kx financial analytics technology tackles Big Data crop research at biotech leader Earlham Institute ...

The Cloud

New hacking technique imperceptibly changes memory virtual servers ...