All Activity

How does it decide how the probability maps should be altered to get a better result?

I don't have a lot of time, so. I will try to be quick (given I have only skimmed the posts since OME asked the question). The algorithm is not predictive in the sense of predicting the future. The predict what you meant by what you typed, and whether or not their response will make sense. For example, if you asked how many legs a dog normally has, it will predict that 4 is the correct answer; not that my some genetic mutation caused by increasing radiation in the atmosphere, they will grow 5 legs in the future. Well, not unless you ask it, anyway. But even if you did, it would use existing data to try and predict the correct answer based on the data it has, not to perform its own scientific analysis and determine for itself. It may take the factors and apply it to an existing theory to mimic scientific analysis and intelligent conclusion drawing, but it is not doing that itself. In terms of AI implementation, it is likely to be a lower level language such as C or C++ = not python because it is too slow; the end point AI functions can be done in Python, but that is on a precomputed large dataset - usually. The algorithm will be surprising small.. usual Bayes theorem or something similar. It is able to be repeated over billions of nodes - or data points. The code will likely be recursive in building the precomputed data set to apply simpler probability maps across the data you provide. And it will likely be recursive in using that map to answer your question. By recursion we mean the same code calls itself.. So, you will be able to crawl heterogenous data sets, over and over using the same code developing deep nested stacks (well, heaps to be accurate) of responses from which it will map as many of the different types of data that can be used to probably represent what ever bit of data it actually is. Tagging data is old these days.. That is what these maps do. This of something like if you type the number 4 into your LLM. That is easy, because it is represented by whatever coding mechanism you are using... and there are only a handful of coding mechanisms - ASCII, EBCDIC (not used anymore) UTF-8, UTF-16 and a few others. But, they are translated to the discrete representation of 4 to AI. No thinking required. Now, say you provide it a scanned doc, which contains the number 4 somewhere. It has to read all of the pixels and look at the shape of each character to determine if it is a letter, number, whitespace, part of a diagram, etc. Say it worked out it was a digit. Now it was to work out what digit it is? It has to look through its maps/nodes and get the most likely digit it is.. it predicts what digit is is because it spits out a probability index of what it has been taught what that digit could possibly be, across any digit which it may match. Is it a 4, or is it a H.. It also has to take into account font. As you know, a 4 can be represented very differently in different fonts... So, the sort answer to your question is that there will be a set of probability algorithms that will read data, build probability maps, and then apply the same probability algorithms to inputs and compare to what it has learned. Machine learning means it constantly refines its probability maps. The technical implementation is likely to to be a lower level language (a lot of Python maths/stats libraries are written in C++).. and there will be a lot of recursion.

I did a search in case I missed it being posted, but Bill Oddie of Goodies fame (and later a wildlife show presenter) passed away yesterday aged 85.

Yes, that’s why I never won Lotto. To be fair that wasn’t really a good test of the methodology. Lotto results are ultimately random, so there was no long term pattern to them that the algorithm could have been trained on to predict the next draw. Had there been an underlying bias it’s possible that the algorithm could eventually have improved the odds in my favour by picking a set of numbers that more likely than not contained next week’s winning numbers. That was my aim at the time. It didn’t work but the programming was fun. While my finite state machine approach was a far less sophisticated AI model than the technology behind ChatGPT or Claude, the principle of using a data set to train an algorithm to mimic human learning by modifying itself by trial and error until it can produce answers to questions that appear logical and intelligent remains the same. Of course it doesn’t mean that the answers are right.

She must be Russian.

Gee, thanks for that. Much of that sounds like self agrandising word salad. So, LLM A.I. is still just using contextual word and phrase comparisons. But you lose me with predictive aspects. I'd agree that probability might be possible, but predictive results gets too close to seeing into the future. That's impossible. My take on A.I.'s own answer is that it's often limited when it comes to initiative or sound logic.

This what it said when I asked it, but don't ask me to explain it: What does A.I. convert human language into, in order to compare the words with the algorithm? AI converts human language into high-dimensional numerical vectors, often called vector embeddings. These are essentially long strings of decimal numbers representing coordinates in a multi-dimensional mathematical "meaning space How Language is Transformed into Math To process your words, AI follows a strict pipeline: Tokenization: The algorithm breaks text down into small pieces called tokens (which can be words, prefixes, or characters) and assigns each a unique ID number. Vector Embeddings: Each token is mapped to a dense list of numbers. A single word can be represented by thousands of dimensions (e.g., 1,536 to 3,072 separate numbers). Each number acts as a coordinate tracking a specific micro-feature of meaning, syntax, or context. Contextual Math: Words with similar meanings or grammatical rules are mathematically positioned close together in this coordinate system. Because language is reduced to geometry, the AI can perform equations like: Vector("King") - Vector("Man") + Vector("Woman") approximates to Vector("Queen") Is it really "just" database matching? No, it is much more dynamic than a static database lookup. While some AI applications use standard vector databases to fetch similar documents (Retrieval-Augmented Generation or RAG), modern Large Language Models (LLMs) operate on active predictive generation, not simple database matching. The Transformer Architecture: Models use Self-Attention mechanisms to dynamically alter a word's vector depending on the words around it. For instance, the word "bank" will be assigned a completely different vector in the phrase "river bank" compared to "investment bank". Probabilistic Computation: The algorithm doesn't copy-paste text from a memory bank. Instead, it processes the text through billions of internal mathematical nodes (weights) to calculate a probability distribution for the next logical token. Emergent Logic: Stacking these layers of math allows the AI to discover the complex geometric patterns of human logic, enabling it to map out reasoning paths rather than merely retrieving matching records. The only part of this that seems close to what I was saying earlier is the Probabilistic Computation bit. Clearly AI models have made some progress since I last used finite state machines to unsuccessfully predict next week's Lotto numbers in the previous century.

Is your head still spinning?

What does A.I. convert human language into, in order to compare the words with the algorithm? Is it really just language modelling and comparing with databasees of similar language? If so, then it isn't rationally thinking.

Poor dog. Imagine being a dog and living with a human being that has less sense than you.

Sheer madness, saying that.

In theory it should be easier for a good computer program to win chess against even the best humans. A computer can look at all possible combinations of moves on both sides for the next what, 6-8 moves or more? Try doing that yourself... I have difficulty working out the possible combinations for the next 3 moves in just the critical areas.

My new air conditioner is ducted 18kW. Hence the big electric bill since we installed it and have it on heating all day! Previously had ducted bottle gas.

Already back at Blighty. My last few days became a bit manic; apart from the little training sortie, the rest of the time was spent with my mum, culminating in an assessment at Eastern Health on Thursday morning, of which I had to leg it after the family contributions to get to the airport for a Thursday arvo flight. Arrived at Heathrow Friday morning and straight into the office. It made sense as it is a lot closed than home, has great showers and I had spend my allocated holiday. Train ride home in the evening. I am back out late October/early November.

This is Barnaby's dog Stella, the one he was barking at on the TV the other day.

Yes, you should pay an amount related to the roads you use and the actual Mileage on those roads. (User Pays) That's FAIR. Nev

I haven't put in panels as it would bre like putting lipstick on a pig. The house really needs knocking down and rebuilding, but I don't have the money for that. I know I am still paying the high price, but I handle it by paying $50 or $100 each pension day to the electric company, and to gas, water rates, etc. That way I build up a credit at a rate I can deal with, and the bill usually comes out $0 to pay. My bank balance revolves in a standard pattern, with no sudden unexpected dives.

I have a few related issues with my trailer. It has cable braking so is restricted to two tonnes GTW which is the maximum I can legally tow with my older model Rodeo. Later models are legal for three tonne so they can pull that GTW if the trailer has electrical braking and the other required specs. My trailer weighs almost 700kg empty giving it a potential payload of 1.3 tonnes. The problem with that is that my little Isuzu 4 cylinder turbo diesel is good for general driving and lighter loads but is struggling to pull that weight even without hills. About 900kg payload (giving 1.6T GTW) is about all that's practical if I keep away from steep hills. Upgrading to a bigger, more powerful truck is not viable with only intermittent towing needs. It would be good if registration and insurance were only charged on actual road use; then it would be viable to own two vehicles, one for towing and hauling loads, the other for normal use.

My bills are pretty modest. Here is a winter bill (also, we were down 1 panel for the first half of the year due to a faulty microinverter) Here is a Feb bill (more solar)

Heat Pump. The figures below are quite accurate. The thing is, though the solar more than covers it. Here is today's input and out from my system. The orange stripes are power being use and the blue power being generated. Note the very long orange 4 stripes is the hot water heater which has been off for a fortnight because we have been away. Normally, the hot water only requires about 30 minutes a day. Today is a mostly cloudy day Running a 2.5 kW split system air conditioner on heat costs about $0.25 to $0.40 per hour, or roughly $2.00 to $3.20 for an 8-hour day, based on an average electricity rate of $0.35 per kWh. [1, 2, 3] Cost Breakdown Hourly Cost: $0.25 – $0.40 per hour Daily Cost (8 hours): $2.00 – $3.20 per day Electrical Draw: Uses about 0.6 kW to 0.8 kW of input power per hour (the 2.5 kW rating is the thermal heating output, not the electricity consumed). [1, 2, 3]

Here is what my energy costs have done over the last year: Gas Elec Total 2025 June 150.00 183.20 333.20 2025 Jul 155.00 155.45 310.45 2025 Aug 310.00 144.25 454.25 2025 Sep 155.00 165.03 320.03 2025 Oct 155.00 74.95 229.95 2025 Nov 155.00 211.58 366.58 2025 Dec 162.29 162.29 2026 Jan 181.61 181.61 2026 Feb 155.00 181.52 336.52 2026 Mar 160.17 160.17 2026 Apr 228.23 228.23 2026 May 180.00 199.61 379.61 Heat pump air conditioner installed 2026 June 339.32 339.32 2026 Jul 518.70 518.70 Electric Hot Water system installed

A husband walks into Victoria’s Secret to buy his wife a sheer nightgown. The clerk shows him several options ranging from $250 to $500 - the more sheer, the higher the price. He picks the sheerest one for $500 and heads home. He gives it to his wife and asks her to go upstairs, try it on, and model it for him. Upstairs, his wife thinks, "It's so sheer it might as well be nothing! I'll just model naked, return it tomorrow, and keep the $500 refund for myself." She appears naked on the balcony and strikes a pose. The husband looks up and says, "Good grief! For $500, you'd think they'd at least iron it!" He didn't see the punch coming and should be able to see out of at least one eye by the weekend.

What sort of heater do you use, and how long can you run it for on a cold day?

That's NOT what showed up on my checking of the source. PLEASE be More careful. My server just contacted me to change a Password urgently . Nev