Peter Wang's banner

Peter Wang

@BrainsAndTennis • 14,506 subscribers

exploring @offpolicyinc. prev co-founder @tryshortcutai. classically trained neuroscientist

Shorts

I grew up playing tennis competitively, and two years ago, before two kids, I was still pretty good (9 UTR). I’ve been getting back into it, but I wish I had a world-class coach telling me what to work on. But good coaches are expensive, selective, and hard to find, whereas models are cheap, available, and should be already good enough. So clearly, nico and I needed to explore what's possible. We chained open-source models post-trained on tennis. This cut each broadcast down to live play, found and tracked the players, ball, court, and labeled every stroke with its type and a 65-joint skeleton. We then post-processed this data into 30+ biomechanical traits across serves, forehands and backhands, normalized for each player’s build and the camera. This included the goat Novak Djokovic, Carlos Alcaraz, Daniil Medvedev, Frances Tiafoe, Alexander Zverev, Ben Shelton, Taylor Fritz, Nicholas Kyrgios, Andrey Rublev, Jannik Sinner, and Tommy Paul. Using the same pipeline, we then processed phone footage of myself and Richard Pham (ex-D1, used to be >13 UTR) playing. Then we gave Claude all the measurements with a blank context. What impressed me: - Without being told, it found my weaker side (forehand) and Richard’s (backhand) - It joined differences that looked separate into one root cause per stroke - My own diagnosis of my forehand was wrong. Claude’s makes a lot more sense What do you think? Brad Gilbert Paul Annacone Darren Cahill Drop youtube links here if you want Coach Claude to give you tips

I grew up playing tennis competitively, and two years ago, before two kids, I was still pretty good (9 UTR). I’ve been getting back into it, but I wish I had a world-class coach telling me what to work on. But good coaches are expensive, selective, and hard to find, whereas models are cheap, available, and should be already good enough. So clearly, nico and I needed to explore what's possible. We chained open-source models post-trained on tennis. This cut each broadcast down to live play, found and tracked the players, ball, court, and labeled every stroke with its type and a 65-joint skeleton. We then post-processed this data into 30+ biomechanical traits across serves, forehands and backhands, normalized for each player’s build and the camera. This included the goat Novak Djokovic, Carlos Alcaraz, Daniil Medvedev, Frances Tiafoe, Alexander Zverev, Ben Shelton, Taylor Fritz, Nicholas Kyrgios, Andrey Rublev, Jannik Sinner, and Tommy Paul. Using the same pipeline, we then processed phone footage of myself and Richard Pham (ex-D1, used to be >13 UTR) playing. Then we gave Claude all the measurements with a blank context. What impressed me: - Without being told, it found my weaker side (forehand) and Richard’s (backhand) - It joined differences that looked separate into one root cause per stroke - My own diagnosis of my forehand was wrong. Claude’s makes a lot more sense What do you think? Brad Gilbert Paul Annacone Darren Cahill Drop youtube links here if you want Coach Claude to give you tips

124,074 просмотров

We (nico and I) motion-sequenced federer's movements on a tennis court into discrete behavioral syllables and made it interactive Also pulled out some interesting changes to his strokes comparing early vs late career that were not immediately obvious (to me). Didn't know his fh was more closed-stanced as he got older

We (nico and I) motion-sequenced federer's movements on a tennis court into discrete behavioral syllables and made it interactive Also pulled out some interesting changes to his strokes comparing early vs late career that were not immediately obvious (to me). Didn't know his fh was more closed-stanced as he got older

24,936 просмотров

Videos

Anya Rossi

sweetdream.ai

SweetDream.ai•Sponsored•Livecam

Watch Anya Live

Anya is streaming live right now! Join her private show and enjoy exclusive content.

Exclusive private shows

1.2k viewers online

Private Show

Join now for exclusive access

Free preview available • Premium content

I and nico ran the fly connectome and found the group of neurons (hΔH, hΔA, hΔI and hΔG) that could allow the fly to navigate using fast synaptic weight updates, not neural activations. This is fast-weight continual learning in a fly, something current LLMs don't do!

I and nico ran the fly connectome and found the group of neurons (hΔH, hΔA, hΔI and hΔG) that could allow the fly to navigate using fast synaptic weight updates, not neural activations. This is fast-weight continual learning in a fly, something current LLMs don't do!

636,246 просмотров • 1 месяц назад

A composite agent harness composed of system1 (multi-agent jevs) + system2 (astra) out-micros an insane computer in my WC3FT RL env (will open-source soon)

A composite agent harness composed of system1 (multi-agent jevs) + system2 (astra) out-micros an insane computer in my WC3FT RL env (will open-source soon)

188,012 просмотров • 1 месяц назад

labelled a shitton of tennis data. can you tell whos who this was a fun little pit stop for much deeper tennis analysis that ill share later

labelled a shitton of tennis data. can you tell whos who this was a fun little pit stop for much deeper tennis analysis that ill share later

113,482 просмотров • 1 месяц назад

We (nico and I) turned Warcraft 3 Frozen Throne into a real RL env wc3env is a fast, concurrent, and deterministic environment for training and evaluating models/agents Frontier agents do ok in isolated battles but fall apart in full 1v1s due to limitations in spatial reasoning, intelligence, and speed Open source and writeup ↓

We (nico and I) turned Warcraft 3 Frozen Throne into a real RL env wc3env is a fast, concurrent, and deterministic environment for training and evaluating models/agents Frontier agents do ok in isolated battles but fall apart in full 1v1s due to limitations in spatial reasoning, intelligence, and speed Open source and writeup ↓

48,447 просмотров • 1 месяц назад

We (nico and I) motion-sequenced federer's movements on a tennis court into discrete behavioral syllables and made it interactive Also pulled out some interesting changes to his strokes comparing early vs late career that were not immediately obvious (to me). Didn't know his fh was more closed-stanced as he got older

We (nico and I) motion-sequenced federer's movements on a tennis court into discrete behavioral syllables and made it interactive Also pulled out some interesting changes to his strokes comparing early vs late career that were not immediately obvious (to me). Didn't know his fh was more closed-stanced as he got older

24,936 просмотров • 1 месяц назад

Больше нет контента для загрузки