We’re presenting the first AI to solve International Mathematical Olympiad problems at a silver medalist level.🥈 Here’s why this is important. ↓ 🧮 The IMO is the world’s most prestigious annual math competition, presenting six problems to participants involving: geometry, algebra, number theory, and combinatorics. Our hybrid system consists of two breakthrough models: ➕AlphaProof, which can generate complex proofs using Lean, a math programming language. This allows answers to be formally verified for correctness. ➕AlphaGeometry 2: an improved version of our previous system to tackle geometry. Trained on more synthetic data than its predecessor, it managed to solve this year’s IMO Problem 4 within 19 seconds. We’re excited to see how our new system could help accelerate AI-powered mathematics, from quickly completing elements of proofs to eventually discovering new knowledge for us - and unlocking further progress towards AGI. Find out more → https://lnkd.in/d6TUU8_Q
In comparison with LLM (which essentialy simply mimic most probable combinatoon of words in a text) all reinforcement learning techniques are somehow "the next level". Various Alfa-go, Alfa-fold and now Alfa-geometry are trully mind boggling. As we see today hype for LLM at some point we will see true reinforcement learning revolution (unfortunayelly this may be the last thing we will see as human race - revolution in reinforcement learning could mean ultimate AI domination).
sorry but i have already solved all of the physics problems. You were a little too late
Interesting! how do you think this compares to human intuition in problem-solving? And do you see any potential applications beyond pure mathematics?
I think we can say by 2035 AI will do almost all analytical tasks better than human beings. So what will human beings do in the future? This is the golden period to do highly creative things where physicality is involved. My passion is dancing and art. When I was in my 30s I had so much work I couldn't spend enough time ( to be a really world class dancer or artist you must sincerely practice at least 4-5 hours every day for say 20 years or more) on my passion. I wish I were now in my 20s. I would have worked for 2-3 hours per day ( making most optimal use of AI) and spend all my time on my passion. What an extraordinary great time awaits humanity. Unbelievable. Develop very strong interest in some hobbies/passion. Life will be heavenly. For people who don't do so will waste there life in social media and other addiction ( with so much leisure what else will people do other than destructive addictive habits...). Thank you Arnab #artificialintelligence #ai #genai #llm #future
I think this is an area in which AI can and will be massively beneficial. It doesn't get all the hype that LLMs do, but I think this could be just as if not more valuable than LLMs like GPT or llama.
Google DeepMind added your #AI everywhere, now I'm struggling with it from the morning and #colab in the north, #Gemini is everywhere... I like it, I even praised it. 👌 Those who praise Anthropic whom I also love and was involved in the initial testing, remember that corporations bear more responsibility for the consequences than local companies. 🤔
This is a monumental leap for AI! Google's achievement of an AI solving the International Mathematical Olympiad at a silver medalist level is truly groundbreaking. This marks a significant step forward in AI's problem-solving capabilities and opens up exciting possibilities for innovation across various fields. Congratulations to the Google team on this incredible accomplishment. ❤️
It needs to be challeged with some problems a little bit beyond the level of the Olympiad kind of problems to analyze if it really has an understanding of the mathematical concepts and the different approaches that show real innovation when it comes to solving hard mathematical problems.
Director, Enterprise Applications at Essex Property Trust
2moThis is a big deal. It goes to show that AI's analytical capabilities in specific domains can grow exponentially as well. To be clear, this is a somewhat separate sub-field from LLMs like ChatGPT. It is more analogous to AlphaGo, and to a certain extend IBM's Deep Blue. For some context, the International Mathematical Olympiad (IMO) has been held since 1959 and is a internationally respected competition for high school kids. Participants are asked to solve complex math problems with high-school level mathematical tools, i.e. calculus is not required. It is a fairly good approximator of how good a country's K-12 STEM education is in general. The United States, China, Russia, South Korea, and Bulgaria are the teams that have taken home the team gold since 2020.