Programming Not Only By Example

Hila Peleg

Recent years have seen great progress in automated synthesis techniques that can automatically generate code based on some intent expressed by the programmer, but communicating this intent remains a major challenge. When the expressed intent is coarse-grained (for example, restriction on the expected type of an expression), the synthesizer often produces a long list of results for the programmer to choose from, shifting the heavy-lifting to the user. An alternative approach, successfully used in end-user synthesis, is programming by example (PBE), where the user leverages examples to interactively and iteratively refine the intent. However, using only examples is not expressive enough for programmers, who can observe the generated program and refine the intent by directly relating to parts of the generated program.

We present a novel approach to interacting with a synthesizer using a granular interaction model. Our approach employs a rich interaction model where (i) the synthesizer decorates a candidate program with debug information that assists in understanding the program and identifying good or bad parts, and (ii) the user is allowed to provide feedback not only on the expected output of a program but also on the program itself. After identifying a program as (partially) correct or incorrect, the user can also explicitly indicate the good or bad parts, to allow the synthesizer to accept or discard parts of the program instead of discarding the program as a whole.

We show the value of our approach in a controlled user study. Our study shows that participants have a strong preference for granular feedback instead of examples and can provide granular feedback much faster.

Additionally, we present a formal model for interactive synthesis, leveraging an abstract domain of predicates on programs in order to describe the iterative refinement of the specifications and reduction of the candidate program space. We use this model to describe the behavior of several real-world synthesizers. We present two conditions for termination of a synthesis session, one hinging only on the properties of the available partial specifications, and the other also on the behavior of the user. Finally, we show conditions for realizability of the user's intent, and show the limitations of backtracking when it is apparent a session will fail.