QAI (Quantum Artificial Intelligence)
Technology Stack Includes:
Affective Neuron™

(View published PCT patent application)

Affective Neuron™, designed for hi-performance hardware implementation and also software, is a reinforcement learning neural network technology that, when applied, is conditioned not trained, is unbiased and learns in real-time, continuously.

Machine learning algorithms have not changed in several decades. What’s changed is the hardware (speed) that processes the algorithms. AI is not a software challenge, it’s all about hardware. intends to fundamentally impact the limitations of artificial intelligence with novel breakthroughs on both the hardware and software fronts. Much more to come, so stay tuned.

~ ~ ~

Connectionist models of cognition, which are pervasive, are fundamentally flawed. Brain neurons are not simple processing elements, they are complex analog systems.

Current silicon hardware technologies are a dead end, limited to 4 GHz performance. Not a path forward to robust unbiased machine learning let alone the singularity.

Unlike current neural network machine learning technologies, Affective Neuron™ is not based upon the connectionist model of cognitive processing. The connectionist model is over 30 years old. The central connectionist principle is that mental phenomena can be described by interconnected networks of simple and often uniform units.

With AN, the neurons are heterogeneous complex processing elements capable of storing and sharing complete memories.

Recent experiments on brain cells demonstrate that neurons are complex processing elements. This is opposite to the connectionist model of cognition. ANs are interconnected through novel dynamic quantum diffusion processing networks, which mimics how biological neuron synaptic interfaces function.