1: \begin{abstract}

2: Huffman Codes are {\em optimal}  Instantaneous Fixed-to-Variable (FV) codes  in which  every source symbol can only be encoded by one codeword. Relaxing these constraints permits constructing better FV codes.

3: More specifically, recent work has shown that AIFV-$m$ codes can beat Huffman coding.  AIFV-$m$ codes construct am $m$-tuple of different coding trees between which the  code alternates  and  are only {\em almost instantaneous} (AI).  This means that decoding  a word might require a delay of a finite number of bits.

4:  

5: Current algorithms for constructing optimal AIFV-$m$ codes are iterative processes that construct progressively ``better sets'' of code trees.    

6: The  processes have been proven to finitely converge to the optimal code but with no known bounds  on the convergence rate.

7: 

8: This paper derives  a geometric interpretation  of the space of binary AIFV-$2$ codes, permitting the development of the first  polynomially time-bounded iterative procedures for constructing optimal AIFV   codes.

9:  

10: We first show that a simple binary search procedure can replace the current iterative process to construct  optimal binary AIFV-$2$ codes. We then 

11: describe how to frame the problem as  a linear programming with an exponential number of constraints but a polynomial-time  separability oracle.  This permits using the 

12: Gr{\"o}tschel,  Lov{\'a}sz and  Schrijver

13: ellipsoid method to solve the problem in a polynomial number of steps. While more complicated, this second method has the potential to lead to a polynomial time algorithm to construct optimal AIFV-$m$ codes for general $m$.

14: 

15: \medskip

16: 

17: Keywords:  AIFV Codes, AIFV-$m$ Codes. Linear Programming.  Ellipsoid Methods

18: \end{abstract}

19: