BRN Discussion Ongoing
- Thread starter TechGirl
- Start date
Lol, Could you ask for me.
Thanks Violin, yes things are good over here and getting back to near normal. I have also found Tony to be very responsive, patient and a good guy. My last email was sent a couple of months ago and I guess am a little frustrated that I haven't had a reply. From everything that was said towards the end of last year, I really expected the Akida USB to be on the market and I just wanted to know if the plan to release was still in the pipeline. I didn't send a follow up email so cant really blame anyone for the lack of a reply. For all i know my email could have ended up in the junk folder.
Food4 thought
Regular
Use the blood
Good morning chippers.
Free to take for y'all in preparation for another day.
PS: No double dipping please.
as lube.
Stable Genius
Regular
Congrats P.
I’ve been wanting to do that but mines tied up in a government super fund and they won’t let me. I would have doubled my money already if they let me when I first asked. Really peeves me off! Would love to know a work around?
Wags
Regular
Gidday FF, was it Dell??
Violin1
Regular
What? An email from Thailand featuring beer and good times going to the junk folder?!?!?
Vanman1100
Regular
Edge AI and Vision Alliance on LinkedIn: AI Chip Company Syntiant Raises $55 Million to Accelerate Growth
AI Chip Company Syntiant Raises $55 Million to Accelerate Growth
Has anyone done homework on syntiant?
Violin1
Regular
You can set up a separate super fund and do it. My daughter has directed her public service employer super to go to a SMSF and deals with it that way. Downside will be that if it's not a SMSF then you'll probably be limited to 20-25% in THAT new SUPERFUND so would take a while to build. If it's a SMSF then you could put it all in but would of course have to sacrifice a lot to get a decent chunk to invest quickly.
Best of luck.
Zedjack33
Regular
I’m with a gov fund and got mine through about 12 months ago. QSuper.
Proga
Regular
It was shareholder value in the short term ie stick to the strategy for long term growth not go off mission to do dumb things to try and pump the SP in the short term to appease jittery shareholders who misjudged their entry point.
It's up to each individual investor to do their own research and analysis to determine where the company is along its lifecycle and understand the company's growth strategy and timeframes involved to grow shareholder value before investing. However BRN is tricky to judge. It's now out of BRN's hands. They've provided the new technology but it's up to others how fast they can/want to adopt it. It's very much dependant on where they are in their own application lifecycles and new application rollouts and/or whether they can/want to retro fit it or how quickly they understand the new technology and come up with commercial applications they can use it in.
Every shareholder has concerns including myself but they vary. I think BRN is doing a great job atm trying to address some of the concerns by doing a lot of accessible media interviews and their own podcasts. They can't address every concern and please everybody. However, I'm unimpressed with the information in their quarterlies and AGM presentation. The reports are very basic. I've been investing for a long time and have read a few in my time. I think they need to appoint someone who can provide more clarity to write them and include more guidance.
VictorG
Member
Stable Genius
Regular
Food4 thought
Regular
Yes well done, I did 20% last September, October got in for $0.46 c
My super has never looked so good.
I have about 7 years to I retire
So it will be an interesting ride to that day
That with the other 166000 shares I hold
Who knows where we will be in 7 years.
I just wonder if we will look back on this day sp in the small 90’s and go wow I didn’t think we would 1000x it
Enjoy your investment your on the right track.
Slymeat
Move on, nothing to see.
StrategicElements’ battery charges from humidity In the air. Down to almost desert-like levels I believe. Doesn’t need rain.I read up on that battery being charged by human sweat which is certainly a much better idea for plugging into with AKIDA intelligent wearables than needing to carry around a potted plant in a pack on your back. Don’t envy the idea of having to go outside and stand in the rain to keep it wet. It’s bad enough having to put my wife’s indoor plants out every time it rains because they love it.FF
Hi Vanman,
Edge AI and Vision Alliance on LinkedIn: AI Chip Company Syntiant Raises $55 Million to Accelerate Growth
AI Chip Company Syntiant Raises $55 Million to Accelerate Growthwww.linkedin.com
Has anyone done homework on syntiant?
We've looked at Syntiant a couple of times here and "there".
They have a Frankenstein neural arrangement blending analog and digital circuitry.
We looked at Syntiant earlier this (last) year, but they are worth another look.
Many of their patents relate to the computerized design of analog or digital circuits, voice and image recognition.
They started with analog NNs until 2017, progressed through hybrid analog/digital NNs:
US2019034790A1 Systems And Methods For Partial Digital Retraining
https://worldwide.espacenet.com/pat...5/publication/US2019034790A1?q=US2019034790A1
[0007] Disclosed herein is a neuromorphic integrated circuit including, in some embodiments, a multi-layered analog-digital hybrid neural network. The neural network includes a number of analog layers configured to include synaptic weights between neural nodes of the neural network for decision making by the neural network. The neural network also includes at least one digital layer. The digital layer is configured for programmatically compensating for weight drifts of the synaptic weights of the neural network, thereby maintaining integrity of the decision making by the neural network.
[0050] Since the analog multiplier array 200 is an analog circuit, input and output currents can vary in a continuous range instead of simply on or off. This is useful for storing weights of the neural network as opposed to digital bits; however, such weights are subject to weight drifts on account of, for example, electrostatic discharge from cells including transistors programmed with the weights. (See FIG. 4 and accompanying programming description.) In operation of the analog multiplier array 200 , the weights are multiplied by input currents to provide output currents that are combined to arrive at a decision of the neural network. Should the analog layers of the neural network begin to arrive at incorrect decisions (e.g., incorrectly classifying one or more test images of cats to be dogs or incorrectly estimating a person's age from a photograph) on account of weight drifts in the analog layers of the neural network, the digital layer of the neural network can be programmed to correct for the weight drifts allowing the neural network to arrive at correct decisions (e.g., correctly classifying the one or more test images of the cats to be cats or correctly estimating a person's age from a photograph).
[0056] FIG. 5 illustrates a multi-layered hybrid analog-digital neural network 500 in accordance with some embodiments. As shown, the hybrid neural network 500 includes a number of data inputs, a number of analog layers, a digital layer, and a number of data outputs. The number of analog layers is disposed between the number of data inputs and the digital layer. The digital layer is disposed between the number of analog layers and a number of data outputs. Programmable cells including transistors (e.g., cells including transistors M 1 and M 2 of FIG. 4) in the number of analog layers can be programmed with an initial set of weights as set forth herein for one or more classification problems, one or more regression problems, or a combination thereof. During operation of the number of analog layers, which are disposed in an analog multiplier array (e.g., the analog multiplier array 300 ), the weights are multiplied by input currents to provide output currents that are combined to arrive at a decision of the hybrid neural network 500 by means of one or more of the number of data outputs. Decision making for the regression problems includes predicting continuous quantities corresponding to data input into the number of data inputs (e.g., estimating a person's age from a photograph). Decision making for the classification problems includes predicting discrete classes corresponding to data input into the number of data inputs (e.g., classifying an image as an image of a cat or a dog).
###########################################################
PROCESSING MODULES AND METHODS THEREOF
https://worldwide.espacenet.com/pat...8/publication/WO2020028693A1?q=WO2020028693A1
This describes primary and secondary NNs, but they don't describe the circuit layout for individual neuromorphic processing units.
BrainChip uses 2 NNs, a "watchdog" NN, and a classifier NN in its key word spotting patent:
US2017229117A1 LOW POWER NEUROMORPHIC VOICE ACTIVATION SYSTEM AND METHOD
a system and method for controlling a device by recognizing voice commands through a spiking neural network. The system comprises a spiking neural adaptive processor receiving an input stream that is being forwarded from a microphone, a decimation filter and then an artificial cochlea. The spiking neural adaptive processor further comprises a first spiking neural network and a second spiking neural network. The first spiking neural network checks for voice activities in output spikes received from artificial cochlea. If any voice activity is detected, it activates the second spiking neural network and passes the output spike of the artificial cochlea to the second spiking neural network that is further configured to recognize spike patterns indicative of specific voice commands. If the first spiking neural network does not detect any voice activity, it halts the second spiking neural network.
#############################################
Syntiant also work with Renesas:
Advanced voice and image processing capabilities at the edge are provided through a unique
combination of low-power, multi-modal, multi-feature AI inference capabilities
This solution combines Renesas’ RZ/V2M vision AI microprocessor unit (MPU) and Syntiant’s
NDP120 Neural Decision Processor™
############################################
Syntiant Introduces Second Generation NDP120 Deep Learning Processor for Audio and Sensor Apps
https://www.syntiant.com/post/syntiant-introduces-second-generation-ndp120-deep-learning-processor-for-audio-and-sensor-apps
“The NDP120 is the first of a family of semiconductors using our next generation Syntiant Core 2 tensor processor platform that brings performance levels previously found in plugged-in devices to a power level suitable to run on batteries,” said Kurt Busch, CEO of Syntiant. “We took years of real world, low-power edge deep learning experience to develop this architecture into a scalable design optimized to bring neural processing to power constrained deployments.”
The Syntiant® NDP200™ is a special-purpose processor for deep learning and is ideal for always-on applications in battery-powered devices. The NDP200 applies neural processing to run multiple applications simultaneously with minimal battery power consumption. Built using the Syntiant Core 2™ programmable deep learning architecture, NDP200 is designed to natively run deep neural networks (DNN) on a variety of architectures, such as CNN, RNN, and fully connected networks, and it performs vision processing with highly accurate inference at under 1mW. NDP200 brings a level of ML performance that delivers 25x the tensor throughput than the Syntiant Core 1™ found in the Syntiant NDP100™ that are currently shipping in high volume. A programmable Tensilica Hifi3 DSP Is also added for feature extraction and signal processing.
An Easy-to-Use Powerful Neural Network
The Syntiant Core 2 moves larger neural networks into always-on domains with capacity to generate shared embeddings, run ensembles, and other neural architectures concurrently or in cascades. The Syntiant Core 2 is a tensor processing core built from the ground up to support energy efficient inference without compromising ease of programming. This second generation architecture delivers 25x the tensor throughput of the Syntiant Core 1™ found in the Syntiant® NDP100™ and Syntiant® NDP101™ devices that are currently shipping in high volumes.
Graph Native:
The Syntiant Core 2 is built around a highly optimized tensor-based memory and processing system designed to avoid inefficiencies in stored program architectures. Each layer independently controls its parameter, input, and output tensors consistent with graph-based execution, enabling neural designers with full control of multiple concurrent independent networks, or swapping network configurations depending on operating conditions.
Framework Support:
The Syntiant Core 2 training development kit includes tools for running bit-exact simulations directly within high-level modeling languages, such as Tensorflow and Keras. All major frameworks can port to the Syntiant Core 2 runtime, including native support for multiple types of convolutional kernels, kernel striding, kernel dilation, downsampling layers, fully connected layers, pointwise operations, and a variety of activations.
Compression When You Want It:
The evolving practice of neural compression is natively supported by the Syntiant Core 2. Network architectures can mix 1-, 2-, 4-, and 8-bit weights and higher precision bias terms. For the most challenging tasks, the Syntiant Core 2 supports high precision modes, including 16-bit inputs and outputs. Quantize inference when wanted, but not just because the edge processor requires it.
Secret Sauce:
The Syntiant Core 2 supports a variety of task-dependent under the hood optimizations for sparsity and time series that speed up inference and reduce power requirements without extensive post-training network optimization. With design tools supplied by Syntiant, it is possible to explore the complete power, latency, memory, and parameter requirements on the Syntiant Core 2 interactively, and perform large scale hardware-aware hyperparameter searches with the same tools. The Syntiant Core 2 can shorten time-to-product by months or years as compared to more constrained and power-intensive solutions. “Syntiant’s highly configurable deep neural network provides enormous flexibility for low-power audio applications and feature extraction,” said Clark Peng, vice president, head of product management at MSI, a world leader in gaming. “The NDP120’s programmable DSP combined with its highly accurate inference engine are ideal for creating high performance voice command applications that can run across both traditional and machine learning algorithms.”
This article provides an english translation: https://www.enterpriseai.news/2021/01/08/syntiant-aims-new-ai-chip-at-always-on-edge/ Syntiant Aims New AI Chip at ‘Always-On’ Edge January 8, 2021 by George Leopold
Under review as a conference paper at ICLR 2020
HIGH-PERFORMANCE RNNS WITH SPIKING NEURONS
Stable Genius
Regular
Thanks V for the advice.
Sorry for the late reply, missed your post and then whilst discussing it with my wife I just saw it.
I rang my super company a while back and asked them about it and I got a firm no. re moving any money into a smsf.
From memory I would have gotten in at the 40-50c range so it’s dissapointing not to have been able to invest a percent of my own money where I wanted to when I’ve read other have been.
I’ll try again tomorrow.
Cheers
Fact Finder
Top 20
FROM 2017:
www.dell.com
My opinion only DYOR
FF
AKIDA BALLISTA
Dell and Cisco Joint Innovation Drives IT Transformation through Greater Simplicity | Dell
As Cisco Live 2017 kicks off early next week, the Dell Converged Platforms and Solutions Division has exciting updates to share as we continue joint
My opinion only DYOR
FF
AKIDA BALLISTA
Similar threads
