You’re a machine learning specialist who makes neural networks but your cat seems to be better at it. Now you must solve puzzles to build a cat-to-human translation system (who knows what else this cat is capable of!). Earn a fortune, buy kickass cat outfits, learn how machine learning really works!
The core gameplay of this puzzle/simulation game revolves around using a visual programming interface to build "neural networks" capable of solving a variety of real-world problems. Players must carefully select and arrange different algorithmic building blocks, known as "nodes," to create functional systems that can process input data and produce the desired output.
The game presents players with a series of puzzles, each representing a problem that could potentially be solved using machine learning. These puzzles typically involve sorting, classifying, or identifying patterns in sets of geometric shapes (circles, squares, triangles) within a given time limit. As players progress, the puzzles gradually increase in complexity, requiring more sophisticated solutions and a deeper understanding of the underlying concepts.
Solving these puzzles requires players to develop their logical thinking and problem-solving skills. They must carefully analyze each problem, understand the capabilities and limitations of the available tools, and devise efficient strategies to manage resources and meet the time constraints. The game rewards elegance and efficiency, providing a strong sense of satisfaction when players successfully optimize their solutions.
One of the key features of the game is its visual programming interface, which allows players to construct their machine learning systems by dragging and connecting various nodes on the screen. This approach makes the underlying concepts more accessible and intuitive, as players can focus on the high-level logic without needing to write code directly. The visual nature of the interface also contributes to the game's overall aesthetic and atmosphere, with the nodes and connections creating a visually appealing "data flow" as the player's solutions are executed.
While the game is primarily a puzzle experience, it also incorporates educational elements related to machine learning and artificial intelligence. Each node in the visual programming interface is accompanied by a brief description and a link to external resources (e.g., YouTube videos, articles) that provide more detailed information about the corresponding real-world concepts. This approach allows players to explore the foundations of machine learning at their own pace, with the game serving as an introductory springboard to further learning and exploration.
The game offers a good degree of replayability, as players can revisit earlier puzzles to optimize their solutions and earn higher ratings (bronze, silver, gold). Additionally, the game features optional "startup" challenges that introduce more complex and open-ended problems, allowing players to apply their growing knowledge and skills in increasingly sophisticated ways. As players progress, they can also purchase various hardware and software upgrades to improve the performance of their machine learning systems, further encouraging experimentation and refinement of their strategies.