rgupta
Regular
I think the same was declared in last 4c report.
Parsons bought a few, and a few other orders. Because the amount is not too much so in total sales achievement is considered zero against a target of 9 million.
Dyor
BRN Discussion Ongoing
- Thread starter TechGirl
- Start date
Parsons bought a few, and a few other orders. Because the amount is not too much so in total sales achievement is considered zero against a target of 9 million.
Dyor
In terms of material there is no magical number....so with all respect, I disagree with you number of $5M (that you've referenced from here and the other place).
This is a direct cut from the ASX website..
BRN themselves have made price sensitive announcements (Dec 2024), where the values was well below $5M.
You can find further explanations under listing rules if interested.
The fact is that every announcement does not have to price sensitive... BrainChip COULD put out announcements with regards to partnerships etc that are not price sensitive. There is no excuse. Many other companies on the ASX make price sensitive and non price sensitive announcements within ASX guidelines...keeipng shareholders and the market up to date with company progress and going ons.
Whilst we are on announcements, would it be great to see BrainChip put something out about all the testing that Kevin is completing. Non price sensitive, non-ramping, and just explaining what Kevin is doing, and perhaps explaining how this relates to real life scenarios.
Such a missed opportunity IMO.
Further from ASX:: -
Frangipani
Top 20
A recent journal paper by twelve Politecnico di Turino researchers was uploaded to their university website today: “The inNuCE Research Infrastructure and the Neuromorphic MLOps for AIoT prototyping”.
Seven of the co-authors are from the Interuniversity Department of Regional and Urban Studies and Planning (DIST) and the remaining five are from the Department of Control and Computer Engineering (DAUIN).
The inNuCE Research Infrastructure (RI) hardware includes AKD1000.
“A complementary contribution proposed in this paper is the inNuCE Research Infrastructure (inNuCE RI), a two-pillar infrastructure that instantiates NMLOps in practice for neuromorphic AIoT prototyping. The name inNuCE is derived from the Latin in nuce (“in the shell”, “in embryo”), reflecting the mission to enable developers to create draft prototypes rapidly and then transition them to engineered products once feasibility is established. The first pillar is the Laboratory (inNuCE Lab), a physical facility housing event-based sensors, edge devices, and neuromorphic/digital boards for hands-on experimentation. Pillar two is the Heterogeneous Prototyping Platform (inNuCE HPP), a Platform-as-a-Service (PaaS) that virtualizes heterogeneous HW and enforces reproducibility via containerized toolchains (…).
VI. CONCLUSIONS
“This paper presents the NMLOps process, an evolution of MLOps for the integration of neuromorphic technology in AIoT applications, and the inNuCE RI, a research infrastructure that operationalizes NMLOps on a cloud-native, heterogeneous prototyping platform that is tightly coupled with a physical laboratory. This enables end-to-end, reproducible prototyping and benchmarking across neuromorphic and digital substrates, as well as the validation of represen- tative on-edge AIoT applications such as HAR, Braille reading, navigation tracking, and constraint satisfaction problems. By consolidating toolchains, orchestrating heterogeneous HW with Kubernetes and Slurm, and enforcing rigorous versioning of data, models, and artifacts, inNuCE RI lowers adoption barriers and shortens the path from prototype to engineered system. The utilization of standard storage and versioning systems, in conjunction with the complete accessibility of data and tools within containerized environments, ensures the adherence of the service to the FAIR (Findable, Accessible, Interoperable, Reusable) principles. This facilitates the utilization of the platform for scientific studies, where the rigorous management of data is crucial. More broadly, virtual prototyping platforms such as inNuCE RI offer a low- cost, low-risk environment in which to explore designs and deployment options before committing to HW. Beyond neuromorphic workflows, the infrastructure also supports AIoT use cases that target standard and emerging digital technologies (CPU/GPU/TPU, MCU/TinyML, FPGAs, and other accelerators) through the same NMLOps procedures. This reduces development costs, shortens time-to-market, and broadens the scope of feasible heterogeneous AIoT use cases.
Our major achievement is turning NMLOps from a conceptual adaptation of MLOps into an operational practice, with browser workspaces, a multi-board execution backend, and harmonized evaluation, making cross-target compar- isons and iteration routine. In summary, a cloud-native, NMLOps-driven prototyping infrastructure is an effective catalyst for neuromorphic AIoT, as it preserves reproducibility, reduces risk and cost, and makes heterogeneous HW usable for real applications. Future work will evaluate and, where appropriate, implement federation with other complementary research infrastructures, as well as formalizing a broader NEP that embeds NMLOps into service-oriented workflows for multi-stakeholder system integration.”
(…)
(…)
Seven of the co-authors are from the Interuniversity Department of Regional and Urban Studies and Planning (DIST) and the remaining five are from the Department of Control and Computer Engineering (DAUIN).
The inNuCE Research Infrastructure (RI) hardware includes AKD1000.
“A complementary contribution proposed in this paper is the inNuCE Research Infrastructure (inNuCE RI), a two-pillar infrastructure that instantiates NMLOps in practice for neuromorphic AIoT prototyping. The name inNuCE is derived from the Latin in nuce (“in the shell”, “in embryo”), reflecting the mission to enable developers to create draft prototypes rapidly and then transition them to engineered products once feasibility is established. The first pillar is the Laboratory (inNuCE Lab), a physical facility housing event-based sensors, edge devices, and neuromorphic/digital boards for hands-on experimentation. Pillar two is the Heterogeneous Prototyping Platform (inNuCE HPP), a Platform-as-a-Service (PaaS) that virtualizes heterogeneous HW and enforces reproducibility via containerized toolchains (…).
VI. CONCLUSIONS
“This paper presents the NMLOps process, an evolution of MLOps for the integration of neuromorphic technology in AIoT applications, and the inNuCE RI, a research infrastructure that operationalizes NMLOps on a cloud-native, heterogeneous prototyping platform that is tightly coupled with a physical laboratory. This enables end-to-end, reproducible prototyping and benchmarking across neuromorphic and digital substrates, as well as the validation of represen- tative on-edge AIoT applications such as HAR, Braille reading, navigation tracking, and constraint satisfaction problems. By consolidating toolchains, orchestrating heterogeneous HW with Kubernetes and Slurm, and enforcing rigorous versioning of data, models, and artifacts, inNuCE RI lowers adoption barriers and shortens the path from prototype to engineered system. The utilization of standard storage and versioning systems, in conjunction with the complete accessibility of data and tools within containerized environments, ensures the adherence of the service to the FAIR (Findable, Accessible, Interoperable, Reusable) principles. This facilitates the utilization of the platform for scientific studies, where the rigorous management of data is crucial. More broadly, virtual prototyping platforms such as inNuCE RI offer a low- cost, low-risk environment in which to explore designs and deployment options before committing to HW. Beyond neuromorphic workflows, the infrastructure also supports AIoT use cases that target standard and emerging digital technologies (CPU/GPU/TPU, MCU/TinyML, FPGAs, and other accelerators) through the same NMLOps procedures. This reduces development costs, shortens time-to-market, and broadens the scope of feasible heterogeneous AIoT use cases.
Our major achievement is turning NMLOps from a conceptual adaptation of MLOps into an operational practice, with browser workspaces, a multi-board execution backend, and harmonized evaluation, making cross-target compar- isons and iteration routine. In summary, a cloud-native, NMLOps-driven prototyping infrastructure is an effective catalyst for neuromorphic AIoT, as it preserves reproducibility, reduces risk and cost, and makes heterogeneous HW usable for real applications. Future work will evaluate and, where appropriate, implement federation with other complementary research infrastructures, as well as formalizing a broader NEP that embeds NMLOps into service-oriented workflows for multi-stakeholder system integration.”
(…)
(…)
Attachments
Last edited:
A few mentions of brainchip in this article
Food4 thought
Regular
I was of the beliefs that CEO are not allowed to say inflammatory things and it’s against Australian law and come with a punishment of up to 15 years in jail
From the crapper…. Why is Brainchip recently a Nr. 1 topic with AI created stuff?
www.linkedin.com
neuromorphiccore.ai
Nuclear Detection in the Palm of Your Hand - NeuromorphicCore.ai | NeuromorphicCore.AI
𝗡𝘂𝗰𝗹𝗲𝗮𝗿 𝗗𝗲𝘁𝗲𝗰𝘁𝗶𝗼𝗻 𝗶𝗻 𝘁𝗵𝗲 𝗣𝗮𝗹𝗺 𝗼𝗳 𝗬𝗼𝘂𝗿 𝗛𝗮𝗻𝗱 ☢️🧠 How researchers at Sandia National Laboratories used brain-inspired chips to identify radioactive isotopes using less energy than a single blink of a status LED. 🔬 The Breakthrough Handheld radiation detectors often run into what engineers call...
Nuclear Detection in the Palm of Your Hand - NeuromorphicCore.ai
See how Sandia National Laboratories used Intel Loihi 2 and BrainChip Akida to identify nuclear isotopes using 70,000x less energy than conventional processors—proving that brain-inspired "spiking" algorithms like LCA outperform standard AI in messy, real-world security environments.
neuromorphiccore.ai
Similar threads
