Executive summary: The gradient descent algorithm provides a useful analogy for doing good, suggesting an incremental approach rather than drastic changes, while acknowledging the limitations of optimizing for a single best outcome in a complex world.
Key points:
Gradient descent involves taking small steps in the right direction and reassessing, which can be applied to personal choices in doing good.
This approach counters potential failure modes of extreme self-sacrifice or rapid lifestyle changes in pursuit of doing the most good.
The analogy has limitations, including the infinite dimensions of moral choice, the dynamic nature of the world, and uncertainty about the right metric for goodness.
Recognizing these limitations reinforces the need for adaptability, continuous reassessment, and acceptance of multiple valid approaches to doing good.
Viewing doing good as a process rather than a fixed end state can mitigate risks associated with believing only one specific outcome is acceptable.
This comment was auto-generated by the EA Forum Team. Feel free to point out issues with this summary by replying to the comment, andcontact us if you have feedback.
Executive summary: The gradient descent algorithm provides a useful analogy for doing good, suggesting an incremental approach rather than drastic changes, while acknowledging the limitations of optimizing for a single best outcome in a complex world.
Key points:
Gradient descent involves taking small steps in the right direction and reassessing, which can be applied to personal choices in doing good.
This approach counters potential failure modes of extreme self-sacrifice or rapid lifestyle changes in pursuit of doing the most good.
The analogy has limitations, including the infinite dimensions of moral choice, the dynamic nature of the world, and uncertainty about the right metric for goodness.
Recognizing these limitations reinforces the need for adaptability, continuous reassessment, and acceptance of multiple valid approaches to doing good.
Viewing doing good as a process rather than a fixed end state can mitigate risks associated with believing only one specific outcome is acceptable.
This comment was auto-generated by the EA Forum Team. Feel free to point out issues with this summary by replying to the comment, and contact us if you have feedback.