วีดีโอ

@@JP-xm3qf glad you liked it.

Glad it was helpful!

@@musterschnitt did you have a chance to take a look at any of the other videos I this series? Neural Network Primer th-cam.com/play/PLaJCKi8Nk1hzqalT_PL35I9oUTotJGq7a.html.

@@lucidateAI Thanks for your reply, definitely my "homework" every evening this coming week! Just recently touched upon the entire subject of what a modern transformer is. I'm a partner at small media agency, so the topic is obviously more than important to us. And, honestly, fascinating to me, always had a lay interest in language philosophy and neurosciences - and somewhat all this kind of seems to be coming together here. 👍

A background in neurosciences is certainly helpful! At the risk of being that hated teacher that sets more homework here is a Transformers playlist - th-cam.com/play/PLaJCKi8Nk1hwaMUYxJMiM3jTB2o58A6WY.html&si=1dg2RW8Yy9skruVb. And for 'just the facts' - intro to transformers in sixty seconds playlist - th-cam.com/play/PLaJCKi8Nk1hxM3F0E2f2rr5j6wM8JzZZs.html&si=rcnwmbBabF9XM25a Also 'neural networks in sixty seconds' playlist - th-cam.com/play/PLaJCKi8Nk1hxNSMM8FSWCWsScstfutKGn.html&si=UPVG7iDFMd8vrNUs If you get a chance to watch any of them, then I hope you find them useful. Let me know!

@@lucidateAI Neuroscience. it's really just lay knowledge, i.e. lots of books about it. But, yeah, some allegories obviously in the structural design and set-up of digital "neurons", in relation to biological ones. No risk at all, Sir! Here's one who constantly bites off more than he can chew. :) I've marked all those playlists already and will dive in! Hope my non-existent math knowledge won't bring it to a halt too soon. But then again, why not ask a GPT to give me a fresh-up on matrix notation and processing, for example? 😂 Really curious to understand why GPT processing seems to be a "black box", even for the developers, from a certain point. Has it do to with the sheer complexity and size of functions & parameters? Would seem like another allegory to the biological brain, where a singe neuron with dendrites, axon and synapses and electro-chemical signaling is a lot. But to grasp that potential being multiplied by ~16 billion in the cortex alone is impossible. Thanks again for so generously sharing your knowledge in such an accessible way!

Chat models contain the entire context of the conversation - or at least as much as can fit in the model's context window. So it adds the new generated text to the existing conversation chain. Thus the prior conversations and exchanges will influence the future response generated. This video th-cam.com/video/BCabX69KbCA/w-d-xo.html explores this phenomenon in more detail if you are keen to dive deeper.

Looks like a crypto wallet seed phrase. If it is then if I were you I'd keep them a little more secret than you have to date. If not that then perhaps you have stumbled across the menu at your local hipster café - 'I'll have the "finger ski" with a side of "exhaust beauty," please!' Just be careful not to order the "assault summer" - I hear it's a bit too spicy for most people. And whatever you do, don't say them three times in front of a mirror, or you might summon the ghost of a confused lexicographer

Agreed. While OpenAI have not definitively stated that Strawberry uses these techniques (at least not in the papers I’ve seen thus far), they provide some pretty strong hints that they are using these techniques, or at least analogous ones. It should be acknowledged that they are more sophisticated that those presented in this video; but variations on forward chaining and backtracking seem to be at the heart of their reasoning.

They are all defined at 0:20 in the video. “N” is the Normal or Gaussian distribution from statistics. d1 is a formula with inputs of the option strike, the underlying price, the volatility, the interest rate, the dividend yield and time to maturity of the option. d2 is also a formula with inputs of d1 (determined as above) along with the volatility and time to maturity. Once you have values for d1 and d2 you can use them in the formula to calculate the ideal call and put valuation (also shown at 0:20) in the video. You use the normal distribution to operate on d1 and d2, along with the strike price and price of the underlying to come up with the call and put price. As an exercise you can use a spreadsheet or write some python code to implement all of these formulae to come up with option prices for a range of strikes, underlying prices, volatilities, time to expiration etc. if you are new to Black Scholes and options pricing I’d highly recommend doing this for the insights you will gain into time decay, sensitivity to vol etc.

Thank you for your kind words and compliments. Greatly appreciated!

Thanks @may4081. Glad you found the video informative and the content compelling. Greatly appreciated!

@@Mari_Selalu_Berbuat_Kebaikan Agreed

Glad you found it useful! Appreciate the comment. GenAI can add huge value here. Right now you have to ‘roll your own’ but my assumption is that this type of technology will soon be baked into all the popular email apps.

What is it worth to them? What is their next best alternative to what you have built? If you know the answers to these questions then pricing for any product is straightforward. Without answers to these questions you are just guessing.

Nice!

Glad you found it useful.

Glad to hear you found it useful.!

Glad you found it useful! Do you have additional use-cases of your own for this text-to-code (in this case text-to-SQL) approach?

@@lucidateAI absolutely, I am focused on building AI agents that automates the bi dashboard creations, currently working on private equity fund investments space

Glad you enjoyed it!

It’s all good

That is extremely kind of you to say so. I’m glad you are enjoying the materials and finding them useful. Any suggestions for material that Lucidate hasn’t covered yet?

You are welcome. Appreciated. Glad you found it informative and comprehensive.

Thanks. I’m glad you find the content on the channel useful. Best wishes in utilizing AI productively in your job.

Glad you enjoyed it!

You are correct. The classification head on this model is for binary classification. At 14:00 or so in this video one of the suggestions for extending this app is to look at a classification head for multiple classes. This discussion thread on HF has some links that you might find useful discuss.huggingface.co/t/multilabel-classification-using-llms/79671. If you are happy to share your results and experiences I’d live to hear how you get on.

@@lucidateAI seen the discussion. Clearly text-classification is the domain of encoder based models, or BERT-like models. I'll keep Searching. But I'm studying the hugging face tutorials from scratch. I'll revisit this once I have a good grounding of the hugging face pipelines and their intended use from the ground up

Makes sense. I think the HF YT tutorials are excellent. Time well spent.

Hi @CharlesOkwuagwu clearly each model and training set will have its own idiosyncrasies, so it is naturally impossible to say for certain, but you would expect top see bias in the results at inference time where the model performs well on the majority data that it has seen, but poorly for the minority classes. It will also generalise poorly to real-world unbiassed examples as the model has been trained on biassed data that does not reflect the distribution in the real world. Also the performance metrics will likely be skewed - check out this video th-cam.com/video/a2oZwdwo0M0/w-d-xo.html on performance metrics which explains how performance metrics like accuracy can be misleading. In imbalanced datasets, a model can achieve high accuracy by simply predicting the majority class most of the time. This does not mean the model is performing well on all classes. More informative metrics such as precision, recall, F1-score, and the area under the receiver operating characteristic (ROC-AUC) should be used to evaluate the model's performance more effectively.

@@lucidateAI thanks for the response. I have a curated dataset of customer chats, each labeled by a human, a total of 130 classes. The number in each class goes from 5 to over 6000 . Total records over 30,000. I'm trying to see if maybe I should use an LLM to synthesize chat samples where we have less real samples, to get a better balance. Your thoughts🤔💭

I guess the first question to ask is - “Is the statistics of the dataset representative of the real world?” (And I’m certain that you have already posed that question!!). Clearly if it is then there is little value in generating synthetic data. If not then I’d first ask is there a way to get a more representative dataset before generating synthetic data. While generating synthetic data is a common approach, and with the right controls reasonably safe I’m leery of it. The problem is that AI models will find patterns, whether there is a pattern there or not. If there are biases in the synthetic dataset production that introduces some artificial artefacts in the synthetic dataset then the LLM (or frankly any other AI/ML system) will almost always “discover” it. This can massively contaminate performance during inference. In capital markets many firms generate synthetic prices, and unless you are very careful models trained on synthetic prices perform poorly on real world data. Then consider LLMs themselves. Trained on vast corpora of data sourced from the public Internet. At first it is a reasonable assumption that the idioms if language they were learning were genuine human language. As more and more content becomes LLM generated there is clearly a chance that all LLMs learn is “LLM-ese”. So a long-winded answer (but you did ask!) is to use synthetic data as a last resort and be very careful with its construction. Far better to try and source a representative data set if you can. Good luck!

Thanks @BABEENGINEER I use Manim (MAthematics ANIMation) docs.manim.community/en/stable/tutorials/quickstart.html. This video th-cam.com/video/WoittT72pgA/w-d-xo.htmlsi=euGXRqlxBjGuaa1e at 8:51 shows how I’ve fine-tuned an LLM, CodeLlama in this case, to help write the classes to produce the animation a little faster (and in some cases much better!) than I can craft by hand

Did you get a chance to check out manim?

Glad you found it insightful!

Hi @nickwang4777. No it does not.

I’m probably guilty here of a misleading title. In this video the “problems” that are being solved are problems that pertain only to the Big Tech companies themselves. A more representative title might be “Unleashing the Power of Machine Learning: How Big Tech Solves _their own_ Real-World Problems.” But this is perhaps a bit too long. Or have I misunderstood the point you are making?

@@lucidateAI In short - my dog wants out :) "their own" is more fitting. I also think it makes sense to make people realize that AI, ML can actually work, make money and solve problems. On the really grim side: Have you seen Rob Miles' new video? It's worth it... We cannot live in a world where 5 companies hold all the power and money - with no control whatsoever. We cannot implement AI in THIS world in peace. imo Thanks

Thanks. I’ve not yet seen Rob Miles’ new video. I’ll check it out.

github.com/mrspiggot/LuciSummarizationApplication With thanks and apologies. I've just updated the description. Enjoy the repo!

@@lucidateAI No, thank you for the rapid response. You sir just earned another subscriber👍

Thanks! I hope you enjoy the other videos on the channel as much as this one.

How have you got on with the code in the repo? Have you been able to use it as a platform to add your own functionality?

@@lucidateAI❤

Did you miss me?

@@lucidateAI ur video provide a comprehensive knowledge in this field with technical details, and ur viewer must be someone who very passionate in AI

Thanks!

Glad you enjoyed it

Thanks @ricardofernandez2286 for your kind words. I'm glad you enjoyed the video. This particular video is part of this larger playlist -> th-cam.com/play/PLaJCKi8Nk1hwaMUYxJMiM3jTB2o58A6WY.html

You can find a list of all the Lucidate playlists here -> www.youtube.com/@lucidateAI/playlists

Take a look at these as well th-cam.com/play/PLaJCKi8Nk1hzqalT_PL35I9oUTotJGq7a.html&si=cDgVTll8TiWNK4RV and

@@lucidateAI You deserve!! And thank you very much for your comprehensive and fast response. I will certainly look at the playlists you recommended! Best regards!!

I can't wait to hear what you think!

You are welcome! Glad you enjoyed it!

Agreed!

Me too and I think that the potential is going to increase exponentially. Appreciate the comment as well as your membership and subscription. Richard.

Glad you agree with @zengxiliang. I agree too (naturally...). Are there any areas of focus you are interested in? Mine is predominantly capital markets - which is probably evident. But in my consulting business I see interest from a wide variety of industry sectors outside of finance, and I'm always curious and excited to see where people are utilising generative and agentic AI. Appreciate the comment and the support of the channel. Richard

@@lucidateAI Thanks Richard! I work at a pension fund, we are actively exploring applications of LLMs now, your content is very inspiring and helpful!

Glad you are finding the material useful.

You are welcome. Delighted you found the content useful.

Glad you found it useful. Videos 2 and 3 are already complete and should be on general release next week. (Currently they are available to Lucidate members at the VP, MD or CEO levels). I'm just finishing of the LoRA video as I type. That should be out the week after next. Appreciate the support and I hope you found the content insightful.

Thanks @AbdennacerAyeb! Greatly appreciated. I'm glad you enjoyed the video!

You are welcome. Really glad you found it useful.

This video is focused on which models perform the best at generating source code (that is to say Java, C++, python etc.). On the other hand the subject of this video -> Text Summarisation Showdown: Evaluating the Top Large Language Models (LLMs) th-cam.com/video/8r9h4KBLNao/w-d-xo.html is on text generation/translation/summarization etc. Perhaps the other video is more what you are looking for? In either event the key takeaway is that by all means rely on public, published benchmarks. But if you want to evaluate models on your specific use-case (and if I correctly understand your question, I think you do) then it might be worth considering setting up your own tests and your own benchmarks for your own specific evaluation. Clearly there is a trade off here. Setting up custom benchmarks and tests isn’t free. But if you understand how to build AI models, then it isn’t that complex either.

@@lucidateAI I reformulated a bit my inquiry since it was not clear enough. Can you read it again please?

Thanks for the clarification. The challenge with reading 2 or 3 books will be the size of the LLMs context window (the amount of tokens that can be input at once). Solutions to this involve using vector databases - example here -> th-cam.com/video/jP9swextW2o/w-d-xo.html This involves writing Python code and development frameworks like LangChain. You may be an expert at this, in which case I'd recommend some of the latest Llama models and GPT-4. Alternatively you can use Gemini and Claude 3 and feed in sections of the books at a time (up to the token limit of the LLM). These models tend to perform the best when it comes to breaking down complex, university-level subjects. They seem to have a strong grasp of pedagogical principles and can structure explanations in a clear, easy-to-follow manner. That said, I haven't specifically tested having the models read books on pedagogical tools and then applying those techniques. It's an interesting idea though! Given the understanding these advanced models already seem to have, I suspect that focused training on pedagogical methods could further enhance their explanatory abilities. My recommendation would be to experiment with a few different models, providing them with sample content from the books and seeing how well they internalize and apply the techniques. You could evaluate the outputs to determine which model best suits your needs.

You are very welcome. I’m glad you found it insightful. th-cam.com/play/PLaJCKi8Nk1hyvGVZub2Ar7Az57_nKemzX.html&si=JwiUaQ-UojUXoOwA here are some other video explainers on other Prompt Engineering techniques that I hope you find equally informative.

Appreciate that! Thanks! Glad you found this video and other content on the channel insightful.

Sorry @paaabl0, but thanks for leaving a comment. Let me try, if I may, from another angle. The inputs and outputs to LLMs are natural language. Human text. (Yes, literally they are vectors of subword tokens, but I hope you will forgive the abstraction). If you type text into an LLM, you get text out. AutGPT works by using this feature of LLMs and putting an LLM into a loop. As the inputs and outputs are both natural language you can use clever prompts to control and direct this loop. While there are many prompting techniques you can use 'Plan & Execute' as well as 'ReAct' (REasoning & ACTion) are popular choices here. They work by first instructing. the LLM to go through a sequence of steps - such as 1 Question, 2 Thought, 3 Action, 4 Action Input, 5 Observation (repeat previous 5 steps steps until) 6 Thought == 'I now know the answer to the original question', 7 Divulge answer. See an example of this type of Prompt here: Answer the following questions as best you can. You have access to the following tools: {tools} Use the following format: Question: the input question you must answer Thought: you should always think about what to do Action: the action to take, should be one of [{tool_names}] Action Input: the input to the action Observation: the result of the action ... (this Thought/Action/Action Input/Observation can repeat N times) Thought: I now know the final answer Final Answer: the final answer to the original input question Begin! Question: {input} Thought:{agent_scratchpad} This is authored by Harrison Chase, founder of Langchain and you can access it at the LangChain hub under 'hwchase17/react'. This is the heart of AutoGPT (and other similar attempts at AGI). Buy using the 'input is language / output is also language / prompt LLM into a loop where early stages are about thinking and planning, middle stages are about Reasoning and action and final stages are about conclusion and output', you achieve the type of behaviour associated with tools/projects like AutoGPT. Perhaps this different explanation helped a little, perhaps not. Clearly there a a good many great YT sites on AI and I hope one of them is able to answer your questions around AutoGPT better then I'm able. With thanks for you taking the time to comment on the video.