-
Join us as Chief Community Officer
[updated Jan 5, 2025] We are starting a movement to get people dancing, at scale. It involves extraordinary risk and hard work. But if we succeed… What’s uniquely exciting about starting a startup around dance—or art, for that matter—is how pure it is. This isn’t finance or big tech. The careerists chasing money and prestige…
-
Why Dance Needs to Be for Everyone
Dance is everywhere in our culture—in movies, music videos, social media. Yet it’s nowhere in most people’s lives. Not because they’re uninterested, but because we’ve made it inaccessible. We’re building the company that changes this. The Thesis: A Flywheel for Dance My thesis is simple: the more people getting into dance, the more money there…
-
Reflections on Last Night’s Dance Performance
[Copied from a site that I deleted] An experimental piece is never inherently bad—it’s just more likely than not to fall short, simply because of the nature of experimentation. Just as most new businesses fail due to their inherent novelty, an experimental dance piece is also more likely to receive polite silence rather than rousing…
-
Why I stop competing
[This post received zero editing. What you read is what I wrote. I use AI transcription to convert my voice into text that you read.In the era of heavy editing and filtering, I want my voice to be as genuine and honest as possible, Cloest to the source] I figured I will ask this question…
-
[Weekly Post] Reading paper “Sources of richness and ineffability for phenomenally conscious states”
Link to paper: https://arxiv.org/pdf/2302.06403 My interest in this paper originally came from my study of GFlowNet, which is a new deep learning framework proposed by Mila, and is said to be useful in constructing compositional objects. Consciousness can be considered as a compositional object because it appears to be sequential when ideas and thoughts build,…
-
Weekly DL study note: GFlowNet Code Tutorial (completed)
Completed code: https://github.com/gangfang/littlegfn/blob/main/face_generator.ipynb Pre-requisites Flow Networks (Nothing about training yet) IMPORTANT – MAIN IDEA OF GFN The main idea behind GFlowNet is to interpret the DAG as a flow network, and to think of each edge as a pipe through which some amount of water, or particles, flows. We then want to find a flow where, (a) flow is…
-
Weekly study note: GFlowNet Code Tutorial 2022
Flow Networks (Nothing about training yet) IMPORTANT – MAIN IDEA OF GFN The main idea behind GFlowNet is to interpret the DAG as a flow network, and to think of each edge as a pipe through which some amount of water, or particles, flows. We then want to find a flow where, (a) flow is preserved, (b) the flow…
-
Experiments with a 2M-param transformer
Jupyter notebook: https://github.com/gangfang/nanogpt/blob/main/gpt_dev.ipynb
-
Autoencoder, contrarian thought on AGI and GenAI’s impacts on Software engineering, Interpretability
Autoencoder: Contrarian thought on AGI It’s possible that AGI will never arrive. AI practitioners complain about benchmarks being pushed further and further whenever they make progress. This situation makes sense when viewed from the lens of Bayesian framework. The prior changed, which leads to changes in expectations. Inferring from this, I speculate that it is…
-
Subscribe
Subscribed
Already have a WordPress.com account? Log in now.
Gang Fang's Blog 