Driving Dangerously
April 10, 2019
I've written a few times before about the fragility of neural networks, for example in last year's post Fooling Neural Networks. There is an unstated assumption underlying the training of neural networks that the environment is benign, and there is no adversary trying to fool the neural network. Microsoft started on very limited systems like the original PC, and also assumed a benign environment. Excel had macros, and you could do all sorts of things...like read entries in the contact address book...or send an email...all from within a macro that ran automatically when the spreadsheet was opened. What could possibly go wrong?
To read the full article, click here
Related Semiconductor IP
- Flexible Pixel Processor Video IP
- Complex Digital Up Converter
- Bluetooth Low Energy 6.0 Digital IP
- Verification IP for Ultra Ethernet (UEC)
- MIPI SWI3S Manager Core IP
Related Blogs
- What's driving 3D IC design? Do 2D EDA tools need a total overhaul to support 3D design?
- Tabula, Achronix try driving unique design advantages into solidifed FPGA sector
- Safer SoCs for safer driving
- GDDR6 Beyond Graphics : Memory for AI,VR, and Autonomous Driving
Latest Blogs
- CNNs and Transformers: Decoding the Titans of AI
- How is RISC-V’s open and customizable design changing embedded systems?
- Imagination GPUs now support Vulkan 1.4 and Android 16
- From "What-If" to "What-Is": Cadence IP Validation for Silicon Platform Success
- Accelerating RTL Design with Agentic AI: A Multi-Agent LLM-Driven Approach