Aivest is the First Decentralized Fair Launch App for AI Investment Models. For developers, feel free to hop on our Discord and reach out!
-
Pre-trained AI Investment Models - Information AI and investment advisory AI are live at aivestdao.com, a simple frontend for visualization and code for training both regressors and classifiers. A contract implementing a mean-variance optimization and neural network for predicting best portfolio on chain. Feel free to fair launch your own tokenomic of AI investment models!
-
AlphaAI Model - Aivest utilizes AI to comprehensively track users' investment behaviors, including acquiring new knowledge, secondary trading, market sentiment analysis, airdrops, etc. Ultimately, it forms a digital profile within the crypto industry for each user.
-
ZKML - To protect your data and customized model, ZKML enables the validation of private data using public models or verifying private models with public data.
-
Fair Launch Contract of AI Model - An EVM-compatible contract that allows AI prompt triggers for the token launch function of a specific AI model. Token distribution is conducted fairly, without pre-allocating this particular token to a selected group of users.
-
According to the transaction history on the blockchain and the fluctuation of past market prices associated with a wallet address, along with factors such as time weighting (where transactions closer to the present carry more weight), frequency of transactions, range of traded currencies, and changes in transaction amounts, an analysis of a user's investment style is conducted.
-
Based on this analysis, a suitable portfolio combination is recommended to the user, along with quantifiable data such as monthly return, Sharpe ratio, maximum drawdown, and other metrics.
-
Finally, for a more intuitive understanding of the historical performance of this portfolio combination, a line graph comparing its volatility with the index of the cryptocurrency market for the same period is provided.
"Aspect" here refers to helping different users customize various profit and loss risk conditions.
For example: User A accepts a profit and loss range of "-10%, 20%" for this portfolio, and after a day, User A adjusts the range to "-20%, 20%"; or User B further customizes acceptance of the profit and loss range for this portfolio as "-50%, 50%".
-
GPT model - Generative Pre-trained Transformers, commonly known as GPT, are a family of neural network models that uses the transformer architecture and is a key advancement in artificial intelligence (AI) powering generative AI applications such as ChatGPT.
-
Langchain - LangChain is a framework designed to simplify the creation of applications using large language models.
-
Streamlit - Streamlit is an open-source Python framework for machine learning and data science teams.
-
RAG - RAG synergistically merges LLMs' intrinsic knowledge with the vast, dynamic repositories of external databases.
-
PyPortfolioOpt - PyPortfolioOpt is a library that implements portfolio optimization methods, including classical efficient frontier techniques and Black-Litterman allocation.
-
Alpha-GPT - Alpha-GPT is a new paradigm for alpha mining in the realm of quantitative investment, developing a new interactive alpha mining system framework.
-
MarketSenseAI - MarketSenseAI is an innovative AI-driven framework for stock analysis and selection. Utilizing the advanced reasoning capabilities of GPT-4, MarketSenseAI effectively analyzes a diverse range of data, including company news, fundamentals, market dynamics, and macroeconomic information.