- AI Predicts Coding Mistakes Before Developers Make Them
- Gene Editing Is Now Precise Enough to Modify a Single Letter of DNA
- Samsung’s Futuristic Vision of Whiteboards Is Here and It’s Incredible
- Researchers Are Developing a New Cancer Vaccine Using Virus-Like Particles
- A New AI-Powered App Transcribes Your Conversations in Real-Time
Posted: 05 Mar 2018 09:24 AM PST
Ubisoft, the French video game company, debuted a new artificial intelligence (AI) that proactively flags coding bugs at its recent developer conference in Montreal. The tool, named Commit Assistant, is intended to catch mistakes before developers even commit them in a game’s code.
The company fed approximately 10 years’ worth of code into Commit Assistant from across Ubisoft’s library of software, according to a report from Wired U.K. The AI analyzed where previous mistakes were made in the code and what corrections were applied to rectify those mistakes. This type of machine learning enabled Commit Assistant to predict when a programmer could be at risk of introducing a similar bug.
Audiences expect that video games released by major publishers like Ubisoft would be free from bugs. However, eliminating every single bug from these products can be laborious and time-consuming. GamesIndustry.biz reported that the company’s research and development head, Yves Jacquier, said the AI could save programmers 20 percent of their time. Ubisoft also claimed that Commit Assistant can catch six out of 10 bugs accurately.
If developers don’t catch coding mistakes at their inception, it might fall to paid video game testers to point them out post-production. Once they’ve highlighted an issue, it then falls to the development team to figure out what chunk of errant code is responsible — a massive undertaking. According to Wired U.K., Ubisoft said catching an error before the game hits the shelves can absorb as much as 70 per cent of costs of eliminating bugs.
While Ubisoft’s version of the tool is geared toward video games, the same concept could potentially scope out bugs in all kinds of software development projects. Commit Assistant was created in partnership with the University of Concordia, and the school will soon publish academic papers on the underlying methodology, which could help other software developers utilize the same type of machine learning to streamline their coding process.
However, this AI’s success was in part due to the massive amount of data Commit Assistant was able to pore over before offering bug predictions. A multibillion-dollar international company like Ubisoft has enough raw information to train this type of predictive AI effectively, but smaller developers will likely not be able to match the same results.
The post AI Predicts Coding Mistakes Before Developers Make Them appeared first on Futurism.
Posted: 05 Mar 2018 08:54 AM PST
Gene editing has the power to completely reshape our world. It promises everything from fixing the genetic faults that lead to disease, to destroying disease-causing microbes, to improving the nutrition of the foods we eat and even resurrecting extinct species like the wooly mammoth — all largely thanks to the genetic editing tool CRISPR, which has both popularized this work and made it possible. Now, researchers in Japan have developed a new gene editing technique that uses CRISPR alongside a DNA repair system to modify a single DNA base in the human genome, with what the team’s press release calls “absolute precision.”
The new technique, described in the journal Nature Communications, is known as known as MhAX, or Microhomology-Assisted eXcision. This method came from a group of researchers working to better understand how single DNA mutations known as single nucleotide polymorphisms (SNPs) contribute to hereditary disease. To prove that a SNP causes disease, you need to compare genetically matched “twin” cells.
Yet because these twins differ by only a single SNP, they are incredibly difficult to make. MhAX came out of this difficulty, providing a new pathway to create twin cells.
To make these very precise edits, an SNP modification is first inserted alongside a fluorescent reporter gene that helps researchers to identify modified cells. The researchers engineered a duplicate DNA sequence known as a microhomology (hence the technique’s name) on each side of the fluorescent gene, targeting sites for CRISPR to go in and cut DNA. The researchers were then able to use a DNA repair system known as microhomology-mediated end joining (MMEJ) to remove the fluorescent gene. That left only the single-base edit, in the form of an SNP, behind.
The targeting and removal of this singular gene is truly exact. “Our goal is to generate gene editing technologies which improve our understanding of disease mechanisms, and ultimately lead to therapies,” said Dr. Knut Woltjen, who led the research, in the press release. “We’re confident that MhAX will have broad applicability in current human disease research, and beyond.”
The post Gene Editing Is Now Precise Enough to Modify a Single Letter of DNA appeared first on Futurism.
Posted: 05 Mar 2018 07:29 AM PST
This is the future tech that will replace whiteboards as we know it.
The post Samsung’s Futuristic Vision of Whiteboards Is Here and It’s Incredible appeared first on Futurism.
Posted: 04 Mar 2018 01:49 PM PST
Grant-Winning Cancer Vaccine
The National Cancer Institute has awarded a $2.4 million grant for researchers at Michigan State University (MSU) to engineer a cancer vaccine for animals that uses a virus-like particle to trigger anti-cancer immune responses. While the vaccine will initially be tested against cancer cells in animals, the hope is ultimately to develop a treatment for human patients.
Cancer can be tricky to deal with, particularly since cancer cells are able to suppress the immune system by concealing proteins that trigger an immune response. This vaccine gets around this using two major components: First, Qβ particles, which are the virus-like particles that serve as a red flag for the immune system, and second, tumor-associated carbohydrate antigens (TACAs), which are unique structures present on many cancer cells, but not on healthy cells.
When Qβ particles and TACAs are linked, or “conjugated” together, they effectually teach immune cells that anything they find with TACAs is dangerous and should be destroyed. Theoretically, this should build anti-tumor cell immunity, according to Xuefei Huang, the MSU professor who heads the vaccine research.
"There's no better disease fighter than the immune system; drugs and other therapies can leave cancer cells behind, and then there's nothing left to fight recurrence," Huang said in a press release. "Our vaccine would reduce tumor growth and protect the host against tumor development and redevelopment.”
A New Way to Fight Cancer
“If we can further understand the connections between the structural features of Qβ-TACA conjugates and anti-tumor immunity, we can make a sustained impact on cancer vaccine design," Huang elaborated in the press release. The researchers will study the structure of the Qβ particle in order to strategically modify it to reduce the formation of toxic antibodies and boost the desired cancer-killing cells.
This isn’t the only anti-cancer vaccine currently in development. A different attempt to build immunity against cancer uses stem cells for a more personalized treatment. Another similar vaccine that depends on messenger RNA has already started human trials. Indeed, personalized treatments are a growing trend in anti-cancer research, with one vaccine under clinical tests showing considerable efficacy against skin cancer.
The MSU researchers hope to earn similar successes by first treating cancer in animals. “Spontaneous cancers in dogs and cats provide a true test for the cancer vaccine approach,” explained cancer researcher Vilma Yuzbasiyan-Gurkan in the press release. “This is just one example of the many ways that veterinary and human medical research benefit each other.”
The post Researchers Are Developing a New Cancer Vaccine Using Virus-Like Particles appeared first on Futurism.
Posted: 04 Mar 2018 01:20 PM PST
Transcription on the Go
If you have to deal with transcribing interviews as part of your daily work (like we do), you’ll find a welcome partner in the new Otter app. Developed by former employees from Google and a speech-recognition veteran Nuance, Otter is a free service that transcribes speech on the go through the power of artificial intelligence (AI).
Voice transcription services aren’t new. There are a number of apps available out there, sure, but none seem to work like Otter — and we’re not even talking about the AI aspect yet. Most voice-transcription apps that are free aren’t very accurate, and those that work really well are often too expensive. Additionally, none transcribe in “real-time” as Otter does.
AISense, the startup that developed Otter, saw an opportunity here. There was a market ready for Otter to penetrate, as it proved during its launch at the Mobile World Congress this past week. “This is a perfect time,” AISense CEO and founder Sam Liang told CNet.
This app not only has market trends working in its favor, but it also benefited from a ton of work that has been done recently on voice and AI. There are speech recognition algorithms, which most of us are familiar with because of virtual assistants trained to “talk” to us — Apple’s Siri, Amazon’s Alexa, and Google’s creatively named Assistant. In fact, Amazon is supposedly close to developing another “real-time speech translation” service using Alexa.
All of these developments made it possible to design the Otter app, Liang explained. “With AI tech and deep learning in the last few years, the accuracy for speech recognition has improved dramatically. A few years ago, this system wouldn’t be usable,” he told Cnet.
Otter has a rather simple but intuitive approach to voice transcriptions. As soon as you install the app, available for free for both Android and Apple users, it asks you to do a short and long recording — which you start by pressing the app’s mic icon. These become the basis for your “voiceprint” so that Otter can identify you in the recordings you make.
Why does it need to identify you? Well, because Otter’s live transcriptions are ideally separated by each speaker. Also, the raw transcript of a live conversation you’re recording appears almost immediately in front of you. Otter’s AI also automatically puts tags in every recording and transcription you save for easier file management.
Of course, it isn’t flawless. Otter has certain issues with punctuation, which it tends to leave out, and has difficulty working in crowded places or with loud noise in the background. Plus, you can’t transfer audio recordings not done directly using the app.
Still, for those who do interviews, take copious notes during classes or meetings, or would simply like a hands-free way to record their thoughts as text, an app like Otter could make life much easier. After all, who transcribes speech for the fun of it?
Better try it out while it’s still free, though. AISense plans to implement a subscription model to access extra features later on.
The post A New AI-Powered App Transcribes Your Conversations in Real-Time appeared first on Futurism.
|You are subscribed to email updates from Futurism. |
To stop receiving these emails, you may unsubscribe now.
|Email delivery powered by Google|
|Google, 1600 Amphitheatre Parkway, Mountain View, CA 94043, United States|