Darren LI

In time, the ratio of human work lessens, eventually getting closer to full automation even in areas which require a human specialist—i.e., where small mistakes can have disastrous effects—but likely not without a human somewhere in the loop, especially in areas that are particularly unforgiving to mistakes, such as diagnoses, drug prescriptions,... See more

Tasks with soft edges, where there isn't a clear definition of the optimal solution, are more suitable for AI agents. In games, for example, there may be multiple acceptable ways to behave, and users are forgiving of variations. However, tasks with hard edges, where specific actions or results are expected, can be more challenging for agents.... See more

Etsy, ProductHunt and Pinterest (the latter two both A16z portfolio companies) are building solutions from different directions, and a great many people trying to replicate some form of the boutique model online as well, as I suggested here - to create some kind of specialist curation.

无模型方法专注于直接从与环境的交互中找出value function。此类中的算法严重依赖reward 来学习值函数。因此,具有学习reward的功能很重要。 此外,它们通常更易于实现和调整超参数。

What about larger companies? With some caveats, larger companies aren’t worried about going out of business because their Amplitude is too expensive. They might be price-conscious in terms of negotiating a contract in context of their budget, but most SaaS software is just a line item at the end of the day. What matters more is that the solution is... See more