Arithmetic / Range Coding Checkpoint

Reconstruct states, diagnose implementation errors, and identify what the model guarantees.

Get every question correct to pass this checkpoint. You can review feedback, revise your answers, and submit again.

1. A four-bit coder starts at [0,15], with A:1, B:1, EOF:2 in that order. What is the inclusive high endpoint after selecting B, before normalization? Enter one integer.

2. Which interval triggers E3 for a four-bit coder, after checking E1 and E2?

Select one answer.

[2,13] [8,13] [4,10] [0,7]

3. The encoder has three pending E3 bits and next resolves an output bit of 0. Enter the four bits it writes, without spaces.

4. A decoder has low=0, high=15, code=8 and model A:1, B:1, EOF:2. Which symbol is selected?

Select one answer.

A EOF B No symbol: code lies on a gap

5. An encoder and decoder round-trip every test but both serialize bits least-significant first. What exposes their incompatibility with this exercise?

Select one answer.

More round trips using only their own output Removing EOF from empty messages Counting symbols in the decoded output Decoding the specified MSB-first known streams

6. Why must interval products use wider arithmetic than the 32-bit coding state?

Select one answer.

The initial range size is 2^32 and multiplying by counts can exceed 32 bits Every input byte requires a 64-bit codeword EOF has a larger numeric code than 2^32 The decoder needs floating-point probabilities

Auto-advance on pass

Open access learning · Go · Python · C · SQL · No external dependencies

Transparency · Support this project

Join Discord