HAI
Ethereum
HapticAI
Haptic is an innovative platform that uses reinforcement learning techniques designed for training and subsequently upgrading Large Language Models (LLMs) and various AI models. 🤓
Utilizing its distributed human feedback network, Haptic acts as the bridge, seamlessly connecting the AI training process and human cognition. We believe that this human feedback is a necessity for making decentralised AI models competitive to their centralised competitors. We aim to revolutionize the way AI training and optimization are done at scale by focusing on a token driven rewards redistribution model to gather the highest quality of feedback and training data. 🤝
Our objective is straightforward 🚀
There is a large unlock of model improvements possible through RLHF - which has been observed being utilised at university research labs and AI giants like OpenAI and Anthropic. Haptic is being built to ensure the accessibility of these tools and the network of high-quality human feedback providers to any and all AI developers and teams.
The Plan 🪐
The pseudonymous nature of contributors requires us to provide a reputation based on the work completed by contributors to their wallet addresses. The HAI token will be used like a proof of stake mechanism wherein users are ready to lock up tokens to honestly contribute to the reinforcement learning process.
That’s not all 😱
In the upcoming phase of our project, we will leverage the data collected to further refine and enhance our partner LLMs. This phase is centred on user engagement with specific questions that were previously identified as problematic in terms of alignment with user expectations. The users will provide feedback on how various LLMs performed on similar questions, thereby facilitating a comprehensive comparative analysis.
In Haptic's third phase, we’ll be rolling out several new products and initiatives. We’re already in the development phase of building Hierarchical Reinforcement Learning for our complex Large Language Models with memory, while also using inverse reinforcement learning to enhance our reward models. Context-based feedback collection will also be introduced in our AI chatbots and assistants.
The $HAI Token 💎
The HAI token is the core of the incentive model, powering the reinforcement learning engine and bringing high-quality human feedback. Designed after inspiration from DePIN projects across the entire industry, HAI will be distributed to platform contributors fairly and gradually over time. HAI token holders are the net beneficiaries of the token value accrual mechanism.
To ensure that feedback providers and active participants are the biggest beneficiaries of the success of the product - the HAI token has had no Seed round and will be distributed only through a Fair Launch process via the Fjord liquidity bootstrapping pool.
Website: https://www.hapticai.dev/
TG: https://t.me/hapticai
Docs: https://hapticai.gitbook.io/hapticai
Twitter: https://twitter.com/Haptic_AI
