The December 2022 issue of IEEE Spectrum is here!

Close bar

The Neural Network That Remembers

With short-term memory, recurrent neural networks gain some amazing abilities

12 min read
The Neural Network That Remembers
Illustration: John Hersey

“On tap at the brewpub. A nice dark red color with a nice head that left a lot of lace on the glass. Aroma is of raspberries and chocolate. Not much depth to speak of despite consisting of raspberries. The bourbon is pretty subtle as well. I really don’t know that find a flavor this beer tastes like. I would prefer a little more carbonization to come through. It’s pretty drinkable, but I wouldn’t mind if this beer was available.”

Besides the overpowering bouquet of raspberries in this guy’s beer, this review is remarkable for another reason. It was produced by a computer program instructed to hallucinate a review for a “fruit/vegetable beer.” Using a powerful artificial-intelligence tool called a recurrent neural network, the software that produced this passage isn’t even programmed to know what words are, much less to obey the rules of English syntax. Yet, by mining the patterns in reviews from the barflies at BeerAdvocate.com, the program learns how to generate similarly coherent (or incoherent) reviews.

Keep reading...Show less

This article is for IEEE members only. Join IEEE to access our full archive.

Join the world’s largest professional organization devoted to engineering and applied sciences and get access to all of Spectrum’s articles, podcasts, and special reports. Learn more →

If you're already an IEEE member, please sign in to continue reading.

Membership includes:

  • Get unlimited access to IEEE Spectrum content
  • Follow your favorite topics to create a personalized feed of IEEE Spectrum content
  • Save Spectrum articles to read later
  • Network with other technology professionals
  • Establish a professional profile
  • Create a group to share and collaborate on projects
  • Discover IEEE events and activities
  • Join and participate in discussions

Dialing Down a Quantum Compute Glitch by 100,000x

A low-key solution to qubits’ cosmic ray problem

3 min read
Conceptual computer artwork of electronic circuitry contained within spheres against beams of light, representing how data may be controlled and stored in a quantum computer.
Mehau Kulyk/Science Source

The kind of quantum computers that IBM, Google and Amazon are building suffer catastrophic errors roughly once every 10 seconds due to cosmic rays from outer space. Now a new study reveals a way to reduce this error rate by nearly a half-million-fold to less than once per month.

Quantum computers theoretically can find answers to problems no regular computer might ever hope to solve. Their key ingredients, known as quantum bits, or qubits, are linked together by a quantum effect known as entanglement.

Keep Reading ↓Show less
{"imageShortcodeIds":[]}

Deep Learning Gets a Boost From New Reconfigurable Processor

The ReAAP processor allows AI to be faster, more efficient

2 min read
different colored beams of light shooting up
iStock

This article is part of our exclusive IEEE Journal Watch series in partnership with IEEE Xplore.

Deep learning is a critical computing approach that is pushing the boundaries of technology – crunching immense amounts of data and uncovering subtle patterns that humans could never discern on their own. But for optimal performance, deep learning algorithms need to be supported with the right software compiler and hardware combinations. In particular, reconfigurable processors, which allow for flexible use of hardware resources for computing as needed, are key.

Keep Reading ↓Show less

Learn How Global Configuration Management and IBM CLM Work Together

In this presentation we will build the case for component-based requirements management

2 min read

This is a sponsored article brought to you by 321 Gang.

To fully support Requirements Management (RM) best practices, a tool needs to support traceability, versioning, reuse, and Product Line Engineering (PLE). This is especially true when designing large complex systems or systems that follow standards and regulations. Most modern requirement tools do a decent job of capturing requirements and related metadata. Some tools also support rudimentary mechanisms for baselining and traceability capabilities (“linking” requirements). The earlier versions of IBM DOORS Next supported a rich configurable traceability and even a rudimentary form of reuse. DOORS Next became a complete solution for managing requirements a few years ago when IBM invented and implemented Global Configuration Management (GCM) as part of its Engineering Lifecycle Management (ELM, formerly known as Collaborative Lifecycle Management or simply CLM) suite of integrated tools. On the surface, it seems that GCM just provides versioning capability, but it is so much more than that. GCM arms product/system development organizations with support for advanced requirement reuse, traceability that supports versioning, release management and variant management. It is also possible to manage collections of related Application Lifecycle Management (ALM) and Systems Engineering artifacts in a single configuration.

Keep Reading ↓Show less