This is a game changer! (AlphaTensor by DeepMind explained)
#alphatensor #deepmind #ai
Matrix multiplication is the most used mathematical operation in all of science and engineering. Speeding this up has massive consequences. Thus, over the years, this operation has become more and more optimized. A fascinating discovery was made when it was shown that one actually needs less than N^3 multiplication operations to multiply to NxN matrices. DeepMind goes a step further and creates AlphaTensor, a Deep Reinforcement Learning algorithm that plays a single-player game, TensorGame, in order to find even more optimized algorithms for matrix multiplication. And it turns out, there exists a plethora of undiscovered matrix multiplication algorithms, which not only will make everything from computers to smart toasters faster, but also bring new insights into fundamental math and complexity theory.
Sponsor: Assembly AI
Link:
OUTLINE:
0:00 - Intro
1:50 - Sponsor: Assembly AI (link in description)
3:25 - What even is Matrix Multiplication?
6:10 - A very astounding fact
8:45 - Trading multiplications for additions
12:35 - Matrix Multiplication as a Tensor
17:30 - Tensor Decompositions
20:30 - A formal way of finding multiplication algorithms
31:00 - How to formulate this as a game?
39:30 - A brief primer on AlphaZero / MCTS
45:40 - The Results
48:15 - Optimizing for different hardware
52:40 - Expanding fundamental math
53:45 - Summary & Final Comments
Paper:
Title: Discovering faster matrix multiplication algorithms with reinforcement learning
Abstract:
Improving the efficiency of algorithms for fundamental computations can have a widespread impact, as it can affect the overall speed of a large amount of computations. Matrix multiplication is one such primitive task, occurring in many systems—from neural networks to scientific computing routines. The automatic discovery of algorithms using machine learning offers the prospect of reaching beyond human intuition and outperforming the current best human-designed algorithms. However, automating the algorithm discovery procedure is intricate, as the space of possible algorithms is enormous. Here we report a deep reinforcement learning approach based on AlphaZero1 for discovering efficient and provably correct algorithms for the multiplication of arbitrary matrices. Our agent, AlphaTensor, is trained to play a single-player game where the objective is finding tensor decompositions within a finite factor space. AlphaTensor discovered algorithms that outperform the state-of-the-art complexity for many matrix sizes. Particularly relevant is the case of 4 × 4 matrices in a finite field, where AlphaTensor’s algorithm improves on Strassen’s two-level algorithm for the first time, to our knowledge, since its discovery 50 years ago2. We further showcase the flexibility of AlphaTensor through different use-cases: algorithms with state-of-the-art complexity for structured matrix multiplication and improved practical efficiency by optimizing matrix multiplication for runtime on specific hardware. Our results highlight AlphaTensor’s ability to accelerate the process of algorithmic discovery on a range of problems, and to optimize for different criteria.
Authors: Alhussein Fawzi, Matej Balog, Aja Huang, Thomas Hubert, Bernardino Romera-Paredes, Mohammadamin Barekatain, Alexander Novikov, Francisco J. R. Ruiz, Julian Schrittwieser, Grzegorz Swirszcz, David Silver, Demis Hassabis & Pushmeet Kohli
Links:
Homepage:
Merch:
YouTube:
Twitter:
Discord:
LinkedIn:
If you want to support me, the best thing to do is to share out the content :)
If you want to support me financially (completely optional and voluntary, but a lot of people have asked for this):
SubscribeStar:
Patreon:
Bitcoin (BTC): bc1q49lsw3q325tr58ygf8sudx2dqfguclvngvy2cq
Ethereum (ETH): 0x7ad3513E3B8f66799f507Aa7874b1B0eBC7F85e2
Litecoin (LTC): LQW2TRyKYetVC8WjFkhpPhtpbDM4Vw7r9m
Monero (XMR): 4ACL8AGrEo5hAir8A9CeVrW8pEauWvnp1WnSDZxW7tziCDLhZAGsgzhRQABDnFy8yuM9fWJDviJPHKRjV4FWt19CJZN9D4n
13 views
2
1
8 months ago 00:16:15 1
“ THIS VIDEO MAY SHOCK YOU! “ THE MOON IS NOT WHAT YOU THINK!
8 months ago 00:09:01 1
WATCH!! Trudeau Gives UNHINGED SPEECH After Protestors HECKLE him AGAIN!!
8 months ago 00:06:16 1
iphone 14 pro unboxing (silver) 🤍 | setup + camera test 🌷 (𝒂𝒆𝒔𝒕𝒉𝒆𝒕𝒊𝒄)
8 months ago 00:00:48 1
Say hi, Mimi!
8 months ago 00:03:54 1
Newborn Moose Taken By Grizzly Bear || ViralHog
8 months ago 10:09:37 1
Beautiful singing of morning birds by the calm river. 10 Hours of 4K video.
8 months ago 00:10:20 1
Tell Us Your First Love Story | Easy Spanish 271
8 months ago 00:03:04 1
Old God Mordekaiser Skin Spotlight - Pre-Release - PBE Preview - League of Legends
8 months ago 00:16:26 1
DIY Tote Bag Making | Daily use handbag cutting and stitching | Ladies Purse making at home | BAGS
8 months ago 00:09:37 1
Spanish People: Stereotypes vs. Reality | Easy Spanish 216
8 months ago 00:01:43 1
Tiger Ka Message | Tiger 3 | Salman Khan, Katrina Kaif | Maneesh Sharma | YRF Spy Universe
8 months ago 00:12:22 1
UPPER BODY WORKOUT WITH DUMBBELLS (Shoulder-forearms-chest-triceps-back-biceps)
8 months ago 00:55:45 1
Lyon, One of the best cities in France 🇫🇷 Walking Tour - 4K HDR
8 months ago 00:00:17 1
THIS IS ANIME 4K AS NODT TARTAFORAS
8 months ago 00:03:10 1
AOTU Portable Stove | One Minute Review
8 months ago 00:11:11 1
Top 10 ’Dark Future’ predictions that have come TRUE
8 months ago 00:08:14 1
EL Condor Pasa - El Dorado Orchestra 🇵🇪🦅🇵🇪Anthem of the Andes
8 months ago 00:09:50 1
Build of the Week Season 12 - Episode 3 - Jousis’ Tainted Pact Forbidden Rite Pathfinder
8 months ago 00:01:36 1
1 Trillion BPM (beats per minute) Experiment
8 months ago 00:04:46 1
NOVEMBER11 - Leave me now (YouTube Rocks)
8 months ago 00:04:40 1
XANDRIA - My Curse Is My Redemption (Summer 80’s Remix) (Visualizer) | Napalm Records
8 months ago 00:05:33 1
Meat Loaf - I’d Do Anything For Love (But I Won’t Do That) - ZDF 1993