DeepseekerR41's picture

In a Training Loop 🔄

DeepseekerR41 PRO

dodgyQKVattention1

·

AI & ML interests

FOUNDER OF JADEITE AI TECHNOLOGIES Specialising in Reinforcement Learning/Neuroevolutionary AI (Game AI, chess, Go, Poker, etc), Ensemble learning (including deep tabular learning, MLP ensembles and gradient forests), Computer Vision (deep CNNs and ResNet) and Transformers (sparse attention, sliding window attention, fine tuning). Major fan of Claude Shannon

Recent Activity

published an article 4 days ago

Some Brief Notes: On the nature of Differentiable Intelligence and The Pitfalls of a Neural-Symbolic Paradigm in Solving the problem of Blackbox Feature Learning

updated a dataset 8 days ago

dodgyQKVattention1/ZeldaNetv3SyntheticNEODataset

updated a Space 8 days ago

dodgyQKVattention1/ZeldaNetV3

View all activity

Organizations

Articles 1

Article

Some Brief Notes: On the nature of Differentiable Intelligence and The Pitfalls of a Neural-Symbolic Paradigm in Solving the problem of Blackbox Feature Learning

spaces 5

ZeldaNetV3

Shannon Crosspaths Search Tree (SCST)

Simulate and visualize SCST trading strategy performance

Connect4Demo Multi Objective Neuroevolutionary Optimisation with Lineage Specialization MONO LS

Play Connect Four against an AI opponent

ELIZA-Shannon Informational Transformer

Chat with a psychology‑aware ELIZA bot

DEEPFUTUREV0.5

Predict how a variable change affects other factors

models 1

dodgyQKVattention1/ZeldaNetV3

Updated 23 days ago

datasets 1

dodgyQKVattention1/ZeldaNetv3SyntheticNEODataset

Updated 8 days ago • 24