Neural nets as described in many recent articles are very capable at recognizing objects and written and spoken text. But like anything we can build, or even imagine, they have limitations. One problem is that after training, the neural nets we usually encounter are essentially stateless. They can recognize static patterns but… Read More
Tag: bernard murphy
Google, Deep Reasoning and Natural Language Understanding
Understanding natural language is considered a hard problem in artificial intelligence. You could be forgiven for thinking this can’t be right – surely language recognition systems already have this problem mostly solved? If so, you might be confusing recognition with understanding – loosely, recognition is the phonology… Read More
Rebooting EDA
In the 35 years since commercial EDA went mainstream a lot of advances have been made but the fundamental architecture and even the philosophy of tooling have really not advanced at all. Tools are designed around individual tasks – analysis and optimization within a specific domain – under the assumption that variability within… Read More
Google TPU Another Step in a Shifting Semiconductor Landscape
Markets work when consumers of a widget don’t feel there is significant differentiated value in making their own and would rather get lowest possible cost from experienced widget makers who can amortize their investment over high-volume sales to many customers. But that changes when a large consumer finds they can increase differentiated… Read More
DAC 2016 – Register Now
DAC is again going to be in Austin (reason enough to go), from June 6[SUP]th[/SUP]-8[SUP]th[/SUP] for the main event. A number of events caught my eye:
- Monday AM – custom hardware for algorithmic trading. If you want to know more about FinTech (technology for finance) this could be for you
- Another Monday morning session on Linux
Getting Low Power Design Right in Mixed Signal Designs
Mixed-signal design creates all sorts of interesting problems for implementation and verification flows, particularly when it comes to design for low power. We tend to think of mixed-signal as a few blocks like PLLs, ADCs and PHYs on the periphery of the design. Constrain and verify the digital power requirements up to analog … Read More
Deep Learning and Security
Deep learning based on neural nets is most commonly thought of as a very promising approach to scene recognition, text recognition and similar areas. Now there is increasing interest in applying this technology to security objectives. There were a number of papers on this topic in a 2015 BlackHat conference and there are several… Read More
Neural nets for Qualcomm Snapdragon
Neural nets are hot these days. In this forum certainly you can’t swing a cat without hitting multiple articles on the topic – I’ve written some myself. For me there are two reasons for this interest. First, neural nets are amazingly successful at what they do, for example in image recognition where they can beat… Read More
Ecosystem Partnership for Effective Network Hardware Design
When you’re designing a hardware solution to plug into what is arguably the most complex system of all – the Internet – you can’t get away with a little fake traffic to test whether your box is going to do all the right things at the right performance. You have to model realistic voice, video, data and wireless traffic in… Read More
Software-Driven Verification Drives Tight Links between Emulation and Prototyping
I’ve mentioned many times what has become a very common theme in SoC and system verification – it has to be driven by the software because any concept of exhaustively verifying “everything” is neither feasible nor meaningful. Emulation has become a critical component of this flow in validating and regressing… Read More