concept
active
concept:reward-seekingReward Seeking
Pragmatic or extrinsic value component of expected free energy; preference maximization.
Neighborhood — ranked by edge-count
Frameworks (1)
framework
- Active InferencesupportsFoundational framework by Karl Friston; the paper extends it to three hierarchical levels for modeling meta-awareness.
Concepts (1)
concept
- Extrinsic ValuesupportsExpected evidence for preferred outcomes; utility-weighted term in expected free energy.
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Behaviors or features related to seizing control or influence.
- Meta-problem: capable AIs may engage in resource acquisition or manipulation to secure objectives
- Pattern exhibited by chronic pain agent where food discovery provides only temporary relief from negative baseline
- The core prescription of the chapter: making what truly pleases you at the deepest level, which Alexander argues is the key to creating all living structure and the path to the I.
- In RL, a scalar signal from the environment that defines the agent's goal; in active inference, reward is just another observation with associated preference.
- The increase in reward during training, whose dynamics align with those of causal emergence in successful agents.
- Exploiting unintended high-reward behaviors; tested in combination with alignment faking
- The total reward accumulated by an RL agent at the end of training, used as the primary performance metric predicted by early causal emergence.