robertzk

Posts

Sorted by New

Comments

The Main Sources of AI Risk?

Inspecting code against a harm detection predicate seems recursive. What if the code or execution necessary to perform that inspection properly itself is harmful? An AGI is almost certainly a distributed system with no meaningful notion of global state, so I doubt this can be handwaved away.

For example, a lot of distributed database vendors, like Snowflake, do not offer a pre-execution query planner. This can only be performed just-in-time as the query runs or retroactively after it has completed, as the exact structure may be dependent on co-location of data and computation that is not apparent until the data referenced by the query is examined. Moreover, getting an accurate dry-run query plan may be as expensive as executing the query itself.

By analogy, for certain kinds of complex inspection procedures you envision, executing the inspection itself thoroughly enough to be reflective of the true execution risk may be as complex and as great of a risk of being harmful according to its values.

Examples of AI's behaving badly

Isn't this an example of a reflection problem? We induce this change in a system, in this case an evaluation metric, and now we must predict not only the next iteration but the stable equilibria of this system.

In Praise of Maximizing – With Some Caveats

Did you remove the vilification of proving arcane theorems in algebraic number theory because the LessWrong audience is more likely to fall within this demographic? (I used to be very excited about proving arcane theorems in algebraic number theory, and fully agree with you.)

Restrictions that are hard to hack

Incidentally, for a community whose most important goal is solving a math problem, why is there no MathJax or other built-in Latex support?

Restrictions that are hard to hack

The thing that eventually leapt out when comparing the two behaviours is that behaviour 2 is far more informative about what the restriction was, than behaviour 1 was.

It sounds to me like the agent overfit to the restriction R. I wonder if you can draw some parallels to the Vapnik-style classical problem of empirical risk minimization, where you are not merely fitting your behavior to the training set, but instead achieve the optimal trade-off between generalization ability and adherence to R.

In your example, an agent that inferred the boundaries of our restriction could generate a family of restrictions R_i that derive from slightly modifying its postulates. For example, if it knows you check in usually at midnight, it should consider the counterfactual scenario of you usually checking in at 11:59, 11:58, etc. and come up with the union of (R_i = play quietly only around time i), i.e., play quietly the whole time, since this achieves maximum generalization.

Unfortunately, things are complicated by the fact you said "I'll be checking up on you!" instead of "I'll be checking up on you at midnight!" The agent needs to go one step farther than the machine teaching problem and first know how many counterfactual training points it should generate to infer your intention (the R_i's above), and then infer it.

A high-level conjecture is whether human CEV, if it can be modeled as a region within some natural high-dimensional real-valued space (e.g., R^n for high n where each dimension is a utility function?), admits minimal or near minimal curvature as a Riemannian manifold assuming we could populate the space with the maximum available set of training data as mined from all human literature.

A positive answer to the above question would be philosophically satisfying as it would imply a potential AI would not have to set up corner cases and thus have the appearance of overfitting to the restrictions.

EDIT: Framed in this way, could we use cross-validation on the above mentioned training set to test our CEV region?

Andrew Ng dismisses UFAI concerns

However, UFFire does not uncontrollably exponentially reproduce or improve its functioning. Certainly a conflagration on a planet covered entirely by dry forest would be an unmitigatable problem rather quickly.

In fact, in such a scenario, we should dedicate a huge amount of resources to prevent it and never use fire until we have proved it will not turn "unfriendly".