Like many, I marvel the power of ChatGPT and wonder what it can do for me, the business I work in, and the world as a whole. I am still getting up to speed with some of it’s evolving capabilities and working out how it will impact my life (including writing this blog post).
What I have learnt (from Chat GPT) so far is that:
- ChatGPT is a large language model that works by using deep learning algorithms to generate human-like responses to natural language input. ChatGPT is a type of neural network called a transformer.
- The training data for ChatGPT consists of a massive corpus of text from the internet, including websites, books, and other sources. During training, the model is exposed to a large number of examples of human language and is trained to identify patterns and relationships within the text data.
- When a user inputs a question or prompts into ChatGPT, the model uses its knowledge of language patterns to generate a response. The model generates a probability distribution over the possible responses and then selects the response with the highest probability. This process is repeated until a response is generated that the user finds satisfactory.
- In order to improve the quality of its responses, ChatGPT continually learns from its interactions with users. The model adapts and updates its parameters based on the feedback it receives, allowing it to generate more accurate and natural-sounding responses over time.
- Overall, ChatGPT works by leveraging large amounts of text data and sophisticated machine learning algorithms to generate human-like responses to natural language input.
Financial Simplicity is similar to ChatGPT in that:
- Financial Simplicity is a computational model based on machine lead learning algorithms to generate human-like calculations to the application of investment mandates to investment portfolios.
- The training and learning of Financial Simplicity have consisted of a massive assessment of investment business (and generated test cases) and investment portfolios (multi-currency, multi-account, multi-asset class) from around the world. During it’s formulation and testing the algorithms have been exposed to very large numbers of scenarios and these have been used to train the algorithm team to identify and deal with millions of edge cases
- Whenever an investment firm puts scenarios into Financial Simplicity, the algorithm adapts to the specific situation to generate a response. Unlike a probabilistic model though Financial Simplicity adapts the rules to be used to the specific situation
- In order to improve the quality of its responses, Financial Simplicity continually monitors and tests its results and alerts the team should anything be of concern or breach mandate specifics. The model adapts to its given situation based on parameters provided by an investment firm about their way of managing money and client-specific rules.
- Overall, Financial Simplicity has evolved by testing on millions of test cases and uses sophisticated and adaptive algorithms to generate human-like responses to the challenges of monitoring and calculating adjustments to investor-specific investment portfolios
Naturally, these are very different solutions for different challenges, but the similarities are there in the same way that their designed intention is similar – to leverage algorithmic and machine lead learning to help humans do things better and faster.
We, like Chat GPT, also see that the future is the embedding of our algorithmic capabilities into other wealth and investment technology platforms. Contact us if you are interested to investigate how to achieve this.
