If the memory state can be addressed by its content, itisfurthermore called content-addressable.

The Goldilocks phase is reminiscent of "intelligence from starvation" in Darwinian evolution, where resource limitations drivediscoveryofmore efficient solutions.

However, since this decoder effectively approximates thenth derivative of the input vector, it is very sensitive to noise. In our framework, the input is often very noisy, since it corresponds to the converging points of different learning traces. In this section we describe two linear decoders that differ from that in [35] and are more noise-resilient. A.9 and A.10 is crucial for long temporal horizons, since regularization causes the overall magnitude of the recoveredτ-space to decrease asτ increases3. Normalization amends thedecreasing magnitude problem bymaking theτ-space to sum to 1 for everyτ.

In fact, the bound can be improved by considering a data-dependent measure, e.g., Rademacher