@Merlin1A

CS @ UMass Amherst

Joined August 2018
  • Tweets9
  • Following206
  • Followers12
With the new release of Burn, we introduce a new terminal UI dashboard so that users can follow their training with ease without having to connect to any external application 🔥
2
9
40
3,671
George Hotz is so based.
165
549
123
3,366
882,050
Announcing burn-wgpu, the latest cross-platform GPU backend for the burn deep learning framework, leveraging wgpu. This new backend brings Vulkan, DirectX 11/12, Metal, OpenGL, and WebGPU support, enabling you to deploy your models everywhere! burn-rs.github.io/blog/cross…
1
4
1
16
2,878
Feels like we are closer to having fully self driving cars than reliable bluetooth car pairing…
2
1
14
2,176
Hotter take: ML would have advanced faster if another front-end language had been available and widely adopted instead of Python. One that is interactive yet fast & compilable, multithreaded (no GIL), isn't bloated, doesn't care about white spaces,... E.g. Julia or some Lisp.
Bojan Tunguz
@tunguz21 Feb 2023
Hot take: Machine Learning would not have been nearly as advanced now were it not for Python. Python’s two main virtues in the context of ML: 1. Lowering barriers to entry. 2. As a scripting language, it encourages and enables experimental workflow.
297
360
203
3,188
1,800,120
Thought: Excessive marketing is like an arms race and mostly a negative-sum game: If every (US) company halved their marketing budget, no company nor consumer would be worse off, and the savings and freed labor used productively could easily solve world poverty&hunger&more.
US & World Expenses/Needs per Year in Billion US$ (numbers are rounded guesstimates based on multiple sources) $2000B$ [[World Marketing Expenses (1/3 on Ads)]] $1000B$ [[US Marketing Expenses (1/4 on Ads)]] (5% of GDP) $500B$ [[already covered|required/missing]] World Extreme Poverty Needs ($2$/dayĂ—365daysĂ—700M people) $500B$ [[already covered|required/missing]] US Poverty Needs ($35$/dayĂ—365daysĂ—40M people)

ALT US & World Expenses/Needs per Year in Billion US$ (numbers are rounded guesstimates based on multiple sources) $2000B$ [[World Marketing Expenses (1/3 on Ads)]] $1000B$ [[US Marketing Expenses (1/4 on Ads)]] (5% of GDP) $500B$ [[already covered|required/missing]] World Extreme Poverty Needs ($2$/dayĂ—365daysĂ—700M people) $500B$ [[already covered|required/missing]] US Poverty Needs ($35$/dayĂ—365daysĂ—40M people)

2
20
5
48
Here is why @tenstorrent is huge. (1) They are building a stack of software AND hardware tailored for AI workloads. (2) Good balance of general-purpose compute and accelerators. Better than most GPUs/TPUs. (3) Structure is right. Work transparently scales across 10k cores.
1
10
26
Load more