How To Measure ML Model Accuracy

By Bryon Moyer - 01 Apr, 2021 - Comments: 0

Machine learning (ML) is about making predictions about new data based on old data. The quality of any machine-learning algorithm is ultimately determined by the quality of those predictions. However, there is no one universal way to measure that quality across all ML applications, and that has broad implications for the value and usefulness of machine learning. “Every industry, every d... » read more

Hidden Costs In Faster, Low-Power AI Systems

By Ed Sperling - 20 Jan, 2021 - Comments: 1

Chipmakers are building orders of magnitude better performance and energy efficiency into smart devices, but to achieve those goals they also are making tradeoffs that will have far-reaching, long-lasting, and in some cases unknown impacts. Much of this activity is a direct result of pushing intelligence out to the edge, where it is needed to process, sort, and manage massive increases in da... » read more

Neural Networks Without Matrix Math

By Katherine Derbyshire - 17 Sep, 2020 - Comments: 0

The challenge of speeding up AI systems typically means adding more processing elements and pruning the algorithms, but those approaches aren't the only path forward. Almost all commercial machine learning applications depend on artificial neural networks, which are trained using large datasets with a back-propagation algorithm. The network first analyzes a training example, typically assign... » read more

Bridging Math And Engineering In ML

By Ed Sperling - 08 Jan, 2020 - Comments: 0

Steve Roddy, vice president of products for Arm’s Machine Learning Group, examines the intersection of high-level mathematics in the data science used in machine learning within area, speed, and power limitations, and how to bring these two worlds together with the least amount of disruption. » read more

The Cost Of Accuracy

By Brian Bailey - 13 Dec, 2018 - Comments: 0

How accurate does a system need to be, and what are you willing to pay for that accuracy? There are many sources of inaccuracy throughout the development flow of electronic systems, most of which involve complex tradeoffs. Inaccuracy leaves an impact on your design in ways you are not even aware of, hidden by best practices or guard-banding. EDA tools also inject some inaccuracy. As the i... » read more

How Good Is 95% Accuracy?

By Katherine Derbyshire - 28 Dec, 2017 - Comments: 0

Conventional, deterministic computers don’t make mistakes. They execute a predictable series of computations in response to any given input. The input might be mistaken. The logic behind the operations that are performed might be flawed. But the computer will always do exactly what it has been told to do. When unexpected results occur, they can be attributed to the programmer, the system manu... » read more

Power Modeling And Analysis

By Brian Bailey - 27 Sep, 2017 - Comments: 0

Semiconductor Engineering sat down to discuss power modeling and analysis with [getperson id="11489" p_name="Drew Wingard"], CTO at [getentity id="22605" e_name="Sonics"]; [getperson id="11763" comment="Tobias Bjerregaard"], CEO at [getentity id="22908" e_name="Teklatech"]; Vic Kulkarni, vice president and chief strategy officer at [getentity id="22021" e_name="Ansys"]; Andy Ladd, CEO of Baum; ... » read more

Accuracy In Optical Overlay Metrology

By KLA - 18 Jan, 2017 - Comments: 0

By Barak Bringoltz, Tal Marciano, Tal Yaziv, Yaron DeLeeuw, Dana Klein, Yoel Feler, Ido Adam, Evgeni Gurevich, Noga Sella, Ze’ev Lindenfeld, Tom Leviant, Lilach Saltoun, Eltsafon Ashwal, Dror Alumot and Yuval Lamhot, Xindong Gao, James Manka, Bryan Chen, and Mark Wagner. Abstract In this paper we discuss the mechanism by which process variations determine the overlay accuracy of optical m... » read more

Accuracy Is Relative

By Ann Mutschler - 11 Aug, 2016 - Comments: 0

I've been doing some thinking on the concept of accuracy lately in my article, Does Power Analysis Need To Be Accurate? And, I’ve come to the conclusion that there isn’t a single conclusion to be had. Accuracy is very relative to the task at hand, as well as the use case, and the very specific power requirements. And maybe the focus shouldn't be always on the accuracy, but on what the chan... » read more

Standard Time And The Day Of Two Noons

By Brian Bailey - 20 Nov, 2014 - Comments: 2

Many of you know that I have a kind of fascination with dates, times and the way they have changed over time. Everyone measured time by using the sun 150 years ago, and while this may not have been highly accurate – well, who cared? Farmers got up when the sun rose and did their stuff until the sun set. Most people had no need for time and the idea of a watch would have been..well…untimely.... » read more

← Older posts Newer posts →

Knowledge Centers
Entities, people and technologies explored

Shift Left Is The Tip Of The Iceberg

A transformative change is underway for semiconductor design and EDA. New languages, models, and abstractions will need to be created.

by Brian Bailey

Partitioning In The Chiplet Era

Understanding how chiplets interact under different workloads is critical to ensuring signal integrity and optimal performance in heterogeneous designs.

by Ann Mutschler

3.5D: The Great Compromise

Pros and cons of a middle-ground chiplet assembly that combines 2.5D and 3D-IC.

by Ed Sperling

AI’s Role In Chip Design Widens, Drawing In New Startups

Focus is on letting engineers do much more with the same or fewer resources — and less drudgery.

by Karen Heyman

Key Technologies To Extend EUV To 14 Angstroms

Alongside high-NA EUV will be better-performing photoresists, reduced roughness using passivation and etch, and lateral etching to reduce tip-to-tip dimensions.

by Laura Peters

Memory Fundamentals For Engineers

eBook: Nearly everything you need to know about memory, including detailed explanations of the different types of memory; how and where these are used today; what's changing, which memories are successful and which ones might be in the future; and the limitations of each memory type.

by The SE Staff

What Comes After HBM For Chiplets

The standard for high-bandwidth memory limits design freedom at many levels, but that is required for interoperability. What freedoms can be taken from other functions to make chiplets possible?