Master Key is a patent pending DeepMoney invention that emulates a memory process which the human brain uses. In this present AI system for financial markets the Master Key can memorize past price action to make decisions on predicting future price.

Simultaneously while making a prediction to make a decision DeepMoney AI system can empirically explain why it made that decision. These are ground breaking tools essential for risk management for regulation, compliance, audits and understanding how to improve.

The human brain is constantly engaged in prediction, it is a fundamental aspect of how humans can navigate the world, make decisions, and adapt to changing circumstances. Cognitive science fuse memory and prediction, they go hand in hand. 

DeepMoney conceived this Master Key system in 2017. The technical architecture employs a graph based methodology unlike most of the current AI systems that exist presently. To demonstrate the complexity of this topic we asked OpenAI the following question:

ChatGPT question Jan-24

What is your main weakness as an AI chatbot regarding time and memory?

ChatGPTs answer: January 2024

One of the main weaknesses of AI chatbots like me, based on the GPT-3.5 architecture, lies in limitations related to time and memory:

Contextual Understanding Over Time:

While I can maintain some context in a conversation, my ability to remember and understand the context over extended periods is limited. I don't have the ability to retain information about past interactions for very long.

Limited Memory:

I don't have persistent memory between conversations. Each interaction is stateless, meaning I don't remember past conversations unless the context is explicitly provided in the current session. This can affect the depth and continuity of conversations.

Fixed Knowledge Cutoff:

My training only includes data up until January 2022. I'm not aware of events or developments that have occurred since then. Therefore, I may not have the most up-to-date information on certain topics.

Inability to Learn from Specific Users:

I can't learn and adapt based on individual user interactions. I don't personalize responses or remember specific details about users unless it's within the same conversation session.

Response Time:

While I aim to generate responses quickly, there might be delays in processing complex queries or during periods of high usage. Real-time interactions and instant responses can be a challenge in certain situations.

___________

Naturally we do not compete with ChatGPT, we are a niche vertical that specialize in financial markets. Since DeepMoney technology is utilized to trade financial instruments it is a great endowment to have next generation AI time and memory components built into the architecture. The following two diagrams illustrate Master Key.

Market Memory Flowchart

The memory is an experience dataset that can be controlled by the user. If the asset class is a commodity and 25 years of historical market data is pertinent, this can be included.

However if the asset class is crypto and only the past 4 years of data are valid, and the prior 5 years must be excluded, these parameters can be set.

Market Memory overview

The live market data feed is analyzed in real time, when an opportunity occurs the machine will recall from memory similar events that happened on a particular date and time.

Weights are applied to sort the events by importance. The Master Key can generate up to ten episodic similar events, which is far greater than what a human brain can remember.