Simons Institute for the Theory of Computing
@simonsinstitute.bsky.social
The world's leading venue for collaborative research in theoretical computer science. Follow us at http://YouTube.com/SimonsInstitute.
We were delighted to host Simons Institute Industry Day 2025 today, full of lively exchanges and talks by partner companies, program organizers, and postdoctoral-level research fellows.
November 14, 2025 at 3:09 AM
We were delighted to host Simons Institute Industry Day 2025 today, full of lively exchanges and talks by partner companies, program organizers, and postdoctoral-level research fellows.
2/2 "We're used to making guarantees to within the limits of the assumptions of [our] models...How do we now think about using these state-of-the-art techniques in safety critical control?." Claire Tomlin at the Simons Institute's workshop on Theoretical Aspects of Trustworthy AI.
November 12, 2025 at 2:22 PM
2/2 "We're used to making guarantees to within the limits of the assumptions of [our] models...How do we now think about using these state-of-the-art techniques in safety critical control?." Claire Tomlin at the Simons Institute's workshop on Theoretical Aspects of Trustworthy AI.
1/2 "In control and automation, the big change is the advent of foundation models and the potential use of those in automating systems...The challenge is of course for safety." Claire Tomlin of @ucberkeleyofficial.bsky.social on AI for Safety Critical Control. simons.berkeley.edu/talks/claire...
November 12, 2025 at 2:22 PM
1/2 "In control and automation, the big change is the advent of foundation models and the potential use of those in automating systems...The challenge is of course for safety." Claire Tomlin of @ucberkeleyofficial.bsky.social on AI for Safety Critical Control. simons.berkeley.edu/talks/claire...
2/2 Multiplying 2 n x n matrices requires O(n^w) arithmetic operations, where w=3 for the brute force algorithm. Strassen's method was the first big improvement in '69 (w=2.81), followed by two big jumps in the '80s. The world record today is w=2.3714. simons.berkeley.edu/talks/virgin...
October 31, 2025 at 3:49 PM
2/2 Multiplying 2 n x n matrices requires O(n^w) arithmetic operations, where w=3 for the brute force algorithm. Strassen's method was the first big improvement in '69 (w=2.81), followed by two big jumps in the '80s. The world record today is w=2.3714. simons.berkeley.edu/talks/virgin...
1/2 "It seems silly, but it's a very important problem." Virginia Vassilevska Williams (@mit.edu) on the progress in matrix multiplication algorithms during her Richard M. Karp Distinguished Lecture On Matrix Multiplication Algorithms at the Simons Institute. simons.berkeley.edu/talks/virgin...
October 31, 2025 at 3:49 PM
1/2 "It seems silly, but it's a very important problem." Virginia Vassilevska Williams (@mit.edu) on the progress in matrix multiplication algorithms during her Richard M. Karp Distinguished Lecture On Matrix Multiplication Algorithms at the Simons Institute. simons.berkeley.edu/talks/virgin...
Ever wondered about graph learning? Watch Ameya Velingker and Haggai Maron give a masterful introduction at the Simons Institute's workshop on Graph Learning Meets Theoretical Computer Science. Video: simons.berkeley.edu/talks/ameya-...
October 31, 2025 at 6:14 AM
Ever wondered about graph learning? Watch Ameya Velingker and Haggai Maron give a masterful introduction at the Simons Institute's workshop on Graph Learning Meets Theoretical Computer Science. Video: simons.berkeley.edu/talks/ameya-...
October 28, 2025 at 5:59 PM
October 22, 2025 at 6:56 PM
2/2 There are matrix multiplication algorithms that can do better than Strassen's but only for astronomically large matrices, making them impractical, said Oded Schwartz at the Simons Institute's workshop on Complexity and Linear Algebra Boot Camp. Video: simons.berkeley.edu/talks/oded-s...
October 21, 2025 at 11:14 AM
2/2 There are matrix multiplication algorithms that can do better than Strassen's but only for astronomically large matrices, making them impractical, said Oded Schwartz at the Simons Institute's workshop on Complexity and Linear Algebra Boot Camp. Video: simons.berkeley.edu/talks/oded-s...
1/2 Should have paid attention to matrices during linear algebra classes! In 2026, AI will use ~1% of global electricity, of which ~45-90% will be for matrix multiplications, said Oded Schwartz of Hebrew University of Jerusalem at the Simons Institute. simons.berkeley.edu/talks/oded-s...
October 21, 2025 at 11:14 AM
1/2 Should have paid attention to matrices during linear algebra classes! In 2026, AI will use ~1% of global electricity, of which ~45-90% will be for matrix multiplications, said Oded Schwartz of Hebrew University of Jerusalem at the Simons Institute. simons.berkeley.edu/talks/oded-s...
2/2 For multiplying two n x n matrices, the arithmetic complexity of the standard method is of O(n^3); Strassen's method is of O(n^2.81). Prof. Olga Holtz spoke at the Simons Institute's workshop on Complexity and Linear Algebra Boot Camp. Video: simons.berkeley.edu/talks/olga-h...
October 20, 2025 at 5:55 AM
2/2 For multiplying two n x n matrices, the arithmetic complexity of the standard method is of O(n^3); Strassen's method is of O(n^2.81). Prof. Olga Holtz spoke at the Simons Institute's workshop on Complexity and Linear Algebra Boot Camp. Video: simons.berkeley.edu/talks/olga-h...
October 16, 2025 at 3:45 AM
We are currently welcoming applications for postdoctoral-level research fellowships in our Fall 2026 programs. Apply by November 15!
simons.berkeley.edu/research-fel...
simons.berkeley.edu/research-fel...
October 15, 2025 at 12:56 AM
We are currently welcoming applications for postdoctoral-level research fellowships in our Fall 2026 programs. Apply by November 15!
simons.berkeley.edu/research-fel...
simons.berkeley.edu/research-fel...
Party time
October 11, 2025 at 3:30 AM
Party time
2/2 One example showed a 1556x smaller carbon footprint using a better model, GPU, data center PUE and site, said UC Berkeley's David Patterson at the Simons Institute’s workshop on Algorithmic Foundations for Emerging Computing Technologies Boot Camp. simons.berkeley.edu/talks/david-...
October 9, 2025 at 12:52 PM
2/2 One example showed a 1556x smaller carbon footprint using a better model, GPU, data center PUE and site, said UC Berkeley's David Patterson at the Simons Institute’s workshop on Algorithmic Foundations for Emerging Computing Technologies Boot Camp. simons.berkeley.edu/talks/david-...
1/2 Focusing on the 4Ms of AI—model (FLOPs), machine (GPU, mWh / 1.0e12 FLOPs), mechanization (power usage effectiveness) and maps (sites w/ clean energy)—can reduce AI’s carbon footprint, said UC Berkeley's David Patterson, at the Simons Institute. Video: simons.berkeley.edu/talks/david-...
October 9, 2025 at 12:52 PM
1/2 Focusing on the 4Ms of AI—model (FLOPs), machine (GPU, mWh / 1.0e12 FLOPs), mechanization (power usage effectiveness) and maps (sites w/ clean energy)—can reduce AI’s carbon footprint, said UC Berkeley's David Patterson, at the Simons Institute. Video: simons.berkeley.edu/talks/david-...
2/2 Instead, pick the most efficient model, the latest GPU, and an efficient cloud data center in a low carbon location, said David Patterson, at the Simons Institute's workshop on Algorithmic Foundations for Emerging Computing Technologies Boot Camp. Video: simons.berkeley.edu/talks/david-...
October 3, 2025 at 5:43 AM
2/2 Instead, pick the most efficient model, the latest GPU, and an efficient cloud data center in a low carbon location, said David Patterson, at the Simons Institute's workshop on Algorithmic Foundations for Emerging Computing Technologies Boot Camp. Video: simons.berkeley.edu/talks/david-...
1/2 "Bad AI carbon footprint commandments." @ucberkeleyofficial.bsky.social's David Patterson's "don'ts" for reducing AI's carbon footprint: don't pick the biggest model; don't pick an older GPU; don't pick a local data center. He spoke at the Simons Institute. simons.berkeley.edu/talks/david-...
October 3, 2025 at 5:43 AM
1/2 "Bad AI carbon footprint commandments." @ucberkeleyofficial.bsky.social's David Patterson's "don'ts" for reducing AI's carbon footprint: don't pick the biggest model; don't pick an older GPU; don't pick a local data center. He spoke at the Simons Institute. simons.berkeley.edu/talks/david-...
Applications are open for research fellows and senior visitors for the Fall 2026 programs on Spectral Theory Beyond Graphs, and on Pseudorandomness and High-Dimensional Expansion.
simons.berkeley.edu/research-fel...
simons.berkeley.edu/long-term-vi...
simons.berkeley.edu/research-fel...
simons.berkeley.edu/long-term-vi...
October 3, 2025 at 3:59 AM
Applications are open for research fellows and senior visitors for the Fall 2026 programs on Spectral Theory Beyond Graphs, and on Pseudorandomness and High-Dimensional Expansion.
simons.berkeley.edu/research-fel...
simons.berkeley.edu/long-term-vi...
simons.berkeley.edu/research-fel...
simons.berkeley.edu/long-term-vi...
Join us for Mark Embree's Richard M. Karp Distinguished Lecture next Tuesday! Register to attend in person or view the livestream.
simons.berkeley.edu/events/compu...
simons.berkeley.edu/events/compu...
October 3, 2025 at 2:41 AM
Join us for Mark Embree's Richard M. Karp Distinguished Lecture next Tuesday! Register to attend in person or view the livestream.
simons.berkeley.edu/events/compu...
simons.berkeley.edu/events/compu...
Next week at the Simons Institute, a workshop on Linear Systems and Eigenvalue Problems. Join us!
simons.berkeley.edu/workshops/li...
simons.berkeley.edu/workshops/li...
October 3, 2025 at 2:09 AM
Next week at the Simons Institute, a workshop on Linear Systems and Eigenvalue Problems. Join us!
simons.berkeley.edu/workshops/li...
simons.berkeley.edu/workshops/li...
2/2 "Then...apply similar techniques to analyze dolphin vocalization and behavior," said @mit.edu's Mark Hamilton at the Simons Institute workshop on Decoding Communication in Nonhuman Species IV (co-hosted with Project CETI). Video: simons.berkeley.edu/talks/mark-h...
September 30, 2025 at 4:19 PM
2/2 "Then...apply similar techniques to analyze dolphin vocalization and behavior," said @mit.edu's Mark Hamilton at the Simons Institute workshop on Decoding Communication in Nonhuman Species IV (co-hosted with Project CETI). Video: simons.berkeley.edu/talks/mark-h...
1/2 How can one make sense of dolphin vocalizations and behavior? First, "use machine learning to rediscover human words by just watching unlabeled videos of people talking about things," said @mit.edu's Mark Hamilton, speaking at the Simons Institute. Video: simons.berkeley.edu/talks/mark-h...
September 30, 2025 at 4:19 PM
1/2 How can one make sense of dolphin vocalizations and behavior? First, "use machine learning to rediscover human words by just watching unlabeled videos of people talking about things," said @mit.edu's Mark Hamilton, speaking at the Simons Institute. Video: simons.berkeley.edu/talks/mark-h...
Join us Tuesday at 3:30 p.m. PT for the first Karp Lecture of the year. Register to attend in person or view the livestream.
simons.berkeley.edu/events/virtu...
simons.berkeley.edu/events/virtu...
September 29, 2025 at 11:24 PM
Join us Tuesday at 3:30 p.m. PT for the first Karp Lecture of the year. Register to attend in person or view the livestream.
simons.berkeley.edu/events/virtu...
simons.berkeley.edu/events/virtu...
1/2 While in Hawaii during the pandemic, Markus Freitag realized he should take whale language translation seriously. "I really fell in love with whales. You could just lay in the water and you could hear them singing," he said at the Simons Institute. Video: simons.berkeley.edu/talks/markus...
September 29, 2025 at 3:29 PM
1/2 While in Hawaii during the pandemic, Markus Freitag realized he should take whale language translation seriously. "I really fell in love with whales. You could just lay in the water and you could hear them singing," he said at the Simons Institute. Video: simons.berkeley.edu/talks/markus...