New Bug-Hunting Method for Self-Driving Cars
【Summary】DeepXplore is a new debugging method for deep learning algorithms. It has the potential to save lives by making autonomous cars safer.

Microsoft Word crashes before a document is saved, or Skype cuts out during the middle of a conference call. Most software bugs are annoying, but not deadly. That is, unless you're dealing with self-driving cars. Flaws in autonomous vehicle software have the potential to cause accidents and fatalities. That's why researchers are working on a bug detection method, called DeepXplore, to find problems in deep learning algorithms.
Nueral Network Debugging
Deep learning is a form of artificial intelligence (AI) that allows autonomous vehicles to learn as they go. To do this, deep learning uses what's referred to as neural networks, which enable the vehicle to gather information from data images. Deep learning differs from traditional software that follows a fixed, "if this, do that" structure. Because it's so different, deep leaning has unique bugs that require an equally unique detection method.
DeepXplore uses at least three deep learning neural networks to combat bugs. It cross refences the networks with one another for accuracy. If it finds one of the networks has gone astray, it retrains that network so it can fall back in line. Researchers at Columbia University and Lehigh University worked together to create the method.
The DeepXplore Difference
Until DeepXplore, neural network debugging was rather tedious and inaccurate. According to IEE, one method involves researchers feeding test images into the networks until a wrong decision is made. Another, called adversarial testing, creates a series of test images until one is found that causes the network to falter.
But DeepXplore takes a different approach. It creates test images that are most likely to make three or more neural networks make incorrect decisions. For example, it might look for a certain parameter that could lead two networks to identify a vehicle as a car, while the third identifies it as a human.
DeepXplore also activates many neurons and neural network pathways for the best coverage. This process was able to activate 100 percent of network neurons. That's about 30 percent more on average than either other test method used in deep learning.
At this point, DeepXplore isn't guaranteed to find all the bugs in deep learning AI. It is, however, far more comprehensive than previous bug-hunting methods. It also helps get self-driving cars one step closer to public roadways, by making them safer.
-
Tesla Model S Wagon Displayed in Geneva
-
Audi Reveals its Q4 e-tron Concept at the Geneva Motor Show
-
Goodyear Showcases Concept Tire for Flying Cars; 1950's Autonomous Vehicle
-
Kia to Premier Euro-spec Electric e-Soul at the Geneva Motor Show
-
Ford Developing Electric Mustang Crossover That CEO Says Will “Go Like Hell”
-
BMW to Premiere New Plug-in Hybrid Vehicles & Free EV Charging in Geneva
-
Honda Says EVs, Not Brexit, to Blame for Honda Plant Shutdown
-
EVgo Charging Network Powers 88% More EVs Than Last Year
- Volvo Unveils the XC40 Recharge, its First Fully-Electric SUV
- Waymo Exits Austin, Looks to Focus on Programs in Detroit and Phoenix
- Elon Musk Confirms Higher Capacity Battery Packs for the Tesla Model S/X Plaid Variants
- Tesla Rival Lucid Motors Hosts an Official 'Ground-Building' Event for its New $1 Billion Arizona Factory
- Everything We Know About Ford’s Mustang-Inspired Electric Crossover
- Italian Truck Maker Iveco Unveils the ‘Nikola TRE’, an Electric Truck Built in a Joint Venture with Nikola Motor
- Here Are the Top EV Concepts Showcased at Pebble Beach 2019
- Tesla’s Shanghai Factory is Gearing Up to Start Mass Production of the Model 3
- How Will Ford's New 'Mach E' Mustang-inspired e-SUV Fair Against the Tesla Model Y?
- J.D. Power Survey Reveals Automakers Aren’t Doing Enough to Draw Consumers to EVs