How to Assess Technical Feasibility of Machine Learning Projects

TL;DR

To assess the technical feasibility of machine learning projects, utilize external benchmarks, evaluate human level performance, and consider the availability of predictive features. Analyze the project's historical progress alongside whether the data is structured or unstructured, and determine if it addresses a new or existing capability.

Transcript

in the last video you heard about the step of assessing a project for technical feasibility and for value let's take a deeper look at how you can carry out this diligent step to figure out if a project really is feasible and also how valuable it really is let's start with feasibility is this project idea technically feasible before you've started o... Read More

Key Insights

• 📽️ External benchmarks, such as competitor achievements, can provide valuable insights into the technical feasibility of a project.
• ⚾ Human level performance benchmarks can help determine the feasibility of tasks for learning algorithms based on human capabilities.
• 📽️ The availability of predictive features is essential for assessing the technical feasibility of projects, as they indicate the potential success of predicting outcomes accurately.

Questions & Answers

Q: How can external benchmarks help in assessing the technical feasibility of a project?

External benchmarks, such as research literature or competitor achievements, provide a quick sense of whether a project is technically feasible based on successful implementations by others.

Q: How can human level performance benchmarks aid in evaluating project feasibility?

Human level performance benchmarks involve comparing a human's ability to perform a task with the same data as a learning algorithm to determine if the task is feasible for the algorithm.

Q: Why is the availability of predictive features essential in assessing technical feasibility for projects?

Predictive features that show a strong correlation with the target output are crucial for assessing technical feasibility, as they indicate the potential success of the project in predicting outcomes accurately.

Q: How can the history of a project help predict future progress and feasibility?

The rate of progress in a project's history can serve as a predictor for future improvements, allowing teams to estimate the feasibility and success rate of ongoing projects based on past performance.

Summary & Key Takeaways

• Assessing the technical feasibility of a project involves using external benchmarks, such as research literature or competitor achievements, to determine if similar projects have been successful.

• A matrix is presented to evaluate feasibility based on structured versus unstructured data and whether the project is new or existing.

• Project feasibility can also be assessed using human level performance benchmarks, the availability of predictive features, and the project's historical progress.