B01
Renormalization-group flow
When fine-scale degrees of freedom are averaged out, which parameters still predict the same observable? Compare the error at successive scales and identify the range over which the effective model holds.

X.04 / SHARED INSTRUMENTS
What information about a microscopic state survives an average over many particles?
Statistical mechanics relates many microscopic configurations to macroscopic descriptions. Coarse-graining specifies which distinctions are retained and which are discarded. Wilson's scale-based analysis and a physical memory-erasure experiment offer complementary ways to examine that choice. Entropy should therefore be accompanied by the state space, probability distribution, and physical conditions under which it is being calculated or measured.
Consider four binary positions with exactly two ones. There are six arrangements sharing that coarse count. If the description retains only the count, it cannot distinguish those arrangements. Add a second summary, such as whether the ones are adjacent, and inspect how the grouping changes without changing the underlying configurations.
Create two coarse descriptions of the same small system and explain which predictions each can support and which distinctions it loses.
These models isolate a mechanism you can vary. Their assumptions define the part of this topic they can illuminate.
Change one parameter, compare the numerical output, and export the result. The experiment loads when you open it.
Compare symbol frequencies with the ordering that a code can exploit.
The first sweep changes the probability of A while drawing each symbol independently. The other three symbols share the remaining probability.
As one symbol dominates, the observed symbol entropy falls. Huffman codes use short words for frequent symbols; the comparison excludes headers and the codebook.
The second sweep gives all four new draws equal probability and increases repetition. Long runs reward run-length coding, even when a symbol histogram cannot describe the dependence.
Method. The histogram gives an order-zero Shannon entropy. Huffman coding repeatedly joins the two least frequent branches to build a prefix code. The run-length format stores a two-bit symbol and an eight-bit count; counts run from 1 to 255, and a longer run uses another packet.
Kenneth G. Wilson · Scale transformations and critical phenomena; supports coarse-graining and effective descriptions.
Antoine Bérut and coauthors · A one-bit physical memory experiment approaching an erasure dissipation bound.
B01
When fine-scale degrees of freedom are averaged out, which parameters still predict the same observable? Compare the error at successive scales and identify the range over which the effective model holds.
B03
What is being counted in each definition of entropy? Follow the change from messages to physical states and identify the assumptions behind each expression.
B12
Which fine-scale details can be discarded while preserving the behavior measured at a larger scale?
B14
Under what assumptions can a learning problem be expressed as a statistical-mechanical model?