Happens all the time. Problems come quickly when the datasets used for evaluation are not clean, or the evaluation is incorrect - data leakage, problematic imbalance between groups, distribution shifts vs the actual production data. Or people just checking the average performance, but not typical or worst case. Have seen many people run in circles chasing metrics that are meaningless to the task they are supposed to be solving.
Yes. And the same for DS/ML people also, please. The amount of ML people that can meaningfully drill down and actually understand the data is surprisingly low sometimes. Even worse for being able to understand a phenomena _using data_.
Seems like you have identified a potential business need! If some component could simplify the auth similarly, that would presumably be very valuable. Could be an open source project, or a startup (or both).
No they will specify something like dB SPL, A-weighted with Fast time integration. Sometimes referred to as dB(A) SPL.
And measured with a particular method, which specifies things like distance from sources, environmental conditions, etc.
When referring to sound in the physical world, "decibel" mean dB SPL (sound pressure level). Which is defined as the ratio to the smallest perceivable sound pressure level. Unfortunately that is still a bit underspecified, it may be measured with a frequency weighting like A weighting. And then there is the integration time or other temporal aggregation, but that is separate from decibel/log.
When in analog audio, it usually means dbV, relative to a reference voltage.
And in digital audio, usually dBFS - relative to the maximum amplitude that can be represented.
Property based testing is incredibly useful for state-free systems/modules. Especially those that have a wide/complex input space. Simple examples would be a general (de)serialization library for something like JSON.
In Norway, you can get 25% tax break on R&D costs. It can be combined with other softfunding up to 70% for small companies doing research, and 50% for larger companies doing development.
Not necessarily correct the answer! Sometimes it just changes it, insists that it is correct - but it might still be wrong (perhaps now in a new way)...
Strong endorsement! I would like to add that MicroPython can also be extended in C by writing small (or large) modules in C that one can call from Python. There is a module variant that can be installed at runtime (via package manager) called dynamic native modules. This enables extending with native code without having to rebuild or reflash the firmware.
MicroPython can be used either as the top-level entry point, with your C/C++ code being modules (callable from Python). Or you can embed the interpreter into a C/C++ program - see the "embed" port for an example.
I love decision trees. Conceptually simple, computational efficient and giving very good results for a lot of tasks. I especially use them on microcontroller grade system, via emlearn - which converts scikit-learn models to embedded friendly C code.
These articles are a good and pretty comprehensive introduction. I would have loved to have even more examples around the bias/variance trade off for forests, it is a key concept that not all practitioners have integrated.
I work on machine learning applied to sensor data, for understanding physical phenomena. Both time-series models etc for condition monitoring of technical machinery (specifically HVAC in buildings) at https://soundsensing.no
And also running ML inference directly on microcontroller-based sensors, via the open-source project emlearn - https://github.com/emlearn/emlearn
Yeah I got one of those as well. An the older non-S3 version. Fun for developing, very powerful. I use it for developing ML applications for watches etc (emlearn project). Great device for that, but battery life is not its strong point.
It might be just-about acceptable for a smartwatch. But anything the micro takes out of the power budget means less screen and radio time, which does add constraints.
PineTime, based on NRF52, will get you 4-7 days of practical usage.
I agree that homogenization is the likely default outcome. ML models do tend to have a strong tendency to prioritize modelling the most common things well, which makes output in a generative models also biased towards mean/mode. And there are the limitations in representation in the datasets themselves. And also bias in that the developers are primarily English speaking, so that language gets priority.
But I do no see homogenization as inevitable. LLMs do pick up a wide range of speech patterns, both regional, different periods of time , sociolects and styles. And they are pretty darn good at "role-playing", outputting language tailored to a particular role. And this can be configured rather effectively using a session or system prompt. So if everybody picked their voice, we could perhaps get a broadening of outcomes. Maybe I should add "always speak like a refined scholar from the Victorian era, spiced up with 80ies British goth"...
Not in any meaningful way. That is why Microsoft is making a bug press push, to try to convince decision-makers otherwise.
Let's hope our politicians and industry pundits do not fall for it.
Running models locally is starting to get interesting now. Especially the 30B-A3B version seems like a promising direction, though it is still out of reach on 16 GB VRAM (quite accessible).
Hoping for new Nvidia RTX cards with 24/32 GB VRAM. Seems that we might get to GPT4-ish levels within a few years? Which is useful for a bunch of tasks.
What are good, open source, alternatives to VS Code? That are modern IDEs with decent support for frontend, backend, data science, and embedded (possibly via extensions)? That mostly work out of the box, without having to set up and configure NN things.
AlphaProof is among the most relevant methods here. And because it trains by self-play, instead of historical human data - it has a much better chances of being able to solve novel problems, or come up with solutions that humans have not. It did pretty good at the 2024 Olympiad. Will be interesting to see how 2025 goes.