AI assistants will be everywhere // every human routine task can be done
Identifying key patterns of tactics implemented by rival teams, and developing effective responses, lies at the heart of modern football. However, doing so algorithmically remains an open research challenge. To address this unmet need, we propose TacticAI, an AI football tactics assistant developed and evaluated in close collaboration with domain experts from Liverpool FC. We focus on analysing corner kicks, as they offer coaches the most direct opportunities for interventions and improvements. TacticAI incorporates both a predictive and a generative component, allowing the coaches to effectively sample and explore alternative player setups for each corner kick routine and to select those with the highest predicted likelihood of success. We validate TacticAI on a number of relevant benchmark tasks: predicting receivers and shot attempts and recommending player position adjustments. The utility of TacticAI is validated by a qualitative study conducted with football domain experts at Liverpool FC. We show that TacticAI’s model suggestions are not only indistinguishable from real tactics, but also favoured over existing tactics 90% of the time, and that TacticAI offers an effective corner kick retrieval system. TacticAI achieves these results despite the limited availability of gold-standard data, achieving data efficiency through geometric deep learning.
This paper describes TacticAI, an AI system designed to assist coaches and analysts in developing tactics for corner kicks in soccer/football. The key components of TacticAI are:
1. Predictive models for forecasting the receiver of the corner kick and the probability of a shot being taken from the corner kick situation. These are based on graph neural networks that take player positions, velocities, and team information as input.
2. A generative model for recommending adjustments to player positions and velocities to increase or decrease the probability of a shot, conditioned on the initial corner kick setup.
3. The use of geometric deep learning techniques like group convolutions to build equivariance to horizontal/vertical reflections into the neural network architecture, improving data efficiency.
The performance of TacticAI’s components is quantitatively evaluated on a dataset of Premier League corner kicks. A case study is also conducted with expert analysts from Liverpool FC, showing TacticAI can generate realistic tactical adjustments and retrieve similar past corner kick situations in a manner deemed useful by the experts. The integration of predictive and generative models along with the geometric learning allows TacticAI to provide comprehensive analytical support for developing corner kick tactics.
The AI assistant has had 2.3 million conversations, two-thirds of Klarna’s customer service chats
It is doing the equivalent work of 700 full-time agents
It is on par with human agents in regard to customer satisfaction score
It is more accurate in errand resolution, leading to a 25% drop in repeat inquiries
Customers now resolve their errands in less than 2 mins compared to 11 mins previously
It’s available in 23 markets, 24/7 and communicates in more than 35 languages
It’s estimated to drive a $40 million USD in profit improvement to Klarna in 2024
Cisco AI Assistant simplifies tasks by combining generative AI technologies with our unparalleled scope of data, giving you the ability to generate more secure, AI-driven insights that span devices, applications, security, networks, and the internet.
We find that, simply via a sampling-and-voting method, the performance of large language models (LLMs) scales with the number of agents instantiated. Also, this method is orthogonal to existing complicated methods to further enhance LLMs, while the degree of enhancement is correlated to the task difficulty. We conduct comprehensive experiments on a wide range of LLM benchmarks to verify the presence of our finding, and to study the properties that can facilitate its occurrence. Our code is publicly available
The complexity in building AI agents is actually making them work in real world environments! Only 10% is about the LLMs and its reasoning ability The rest is the painful heavy lifting in terms of code, data and memory. Evaluation and ongoing monitoring is an added layer of complexity It’s like a beautiful orchestra when it comes together
