Environmental Bisimulations Research Articles

Environmental Bisimulations for Probabilistic Higher-order Languages

Environmental bisimulations for probabilistic higher-order languages are studied. In contrast with applicative bisimulations, environmental bisimulations are known to be more robust and do not require sophisticated techniques such as Howe’s in the proofs of congruence. As representative calculi, call-by-name and call-by-value λ-calculus, and a (call-by-value) λ-calculus extended with references (i.e., a store) are considered. In each case, full abstraction results are derived for probabilistic environmental similarity and bisimilarity with respect to contextual preorder and contextual equivalence, respectively. Some possible enhancements of the (bi)simulations, as “up-to techniques,” are also presented. Probabilities force a number of modifications to the definition of environmental bisimulations in non-probabilistic languages. Some of these modifications are specific to probabilities, others may be seen as general refinements of environmental bisimulations, applicable also to non-probabilistic languages. Several examples are presented, to illustrate the modifications and the differences.

Environmental bisimulations for probabilistic higher-order languages

Environmental bisimulations for probabilistic higher-order languages are studied. In contrast with applicative bisimulations, environmental bisimulations are known to be more robust and do not require sophisticated techniques such as Howe’s in the proofs of congruence. As representative calculi, call-by-name and call-by-value λ- calculus, and a (call-by-value) λ-calculus extended with references (i.e., a store) are considered. In each case full abstraction results are derived for probabilistic environmental similarity and bisimilarity with respect to contextual preorder and contextual equivalence, respectively. Some possible enhancements of the (bi)simulations, as ‘up-to techniques’, are also presented. Probabilities force a number of modifications to the definition of environmental bisimulations in non-probabilistic languages. Some of these modifications are specific to probabilities, others may be seen as general refinements of environmental bisimulations, applicable also to non-probabilistic languages. Several examples are presented, to illustrate the modifications and the differences.

From Applicative to Environmental Bisimulation

We illuminate important aspects of the semantics of higher-order functions that are common in the presence of local state, exceptions, names and type abstraction via a series of examples that add to those given by Stark. Most importantly we show that any of these language features gives rise to the phenomenon that certain behaviour of higher-order functions can only be observed by providing them with arguments which internally call the functions again. Other examples show the need for the observer to accumulate values received from the program and generate new names. This provides evidence for the necessity of complex conditions for functions in the definition of environmental bisimulation, which deviates in each of these ways from that of applicative bisimulation.

Environmental bisimulations for higher-order languages

Developing a theory of bisimulation in higher-order languages can be hard. Particularly challenging can be: (1) the proof of congruence, as well as enhancements of the bisimulation proof method with “up-to context” techniques, and (2) obtaining definitions and results that scale to languages with different features. To meet these challenges, we present environment{} bisimulations , a form of bisimulation for higher-order languages, and its basic theory. We consider four representative calculi: pure λ-calculi (call-by-name and call-by-value), call-by-value λ-calculus with higher-order store, and then Higher-Order π-calculus. In each case: we present the basic properties of environment bisimilarity, including congruence; we show that it coincides with contextual equivalence; we develop some up-to techniques, including up-to context, as examples of possible enhancements of the associated bisimulation method. Unlike previous approaches (such as applicative bisimulations, logical relations, Sumii-Pierce-Koutavas-Wand), our method does not require induction/indices on evaluation derivation/steps (which may complicate the proofs of congruence, transitivity, and the combination with up-to techniques), or sophisticated methods such as Howe's for proving congruence. It also scales from the pure λ-calculi to the richer calculi with simple congruence proofs.

A bisimulation-like proof method for contextual properties in untyped [formula omitted]-calculus with references and deallocation

We develop a general method for proving properties of programs under arbitrary contexts–including (but not limited to) observational equivalence, space improvement, and a form of memory safety of the programs–in untyped call-by-value λ -calculus with first-class, dynamically allocated, higher-order references and deallocation. The method generalizes Sumii et al.’s environmental bisimulation technique, and gives a sound and complete characterization of each proved property, in the sense that the “bisimilarity” (the largest set satisfying the bisimulation-like conditions) equals the set of terms with the property to be proved. We give examples of contextual properties concerning typical data structures such as linked lists, binary search trees, and directed acyclic graphs with reference counts, all with deletion operations that release memory. This shows the scalability of the environmental approach from contextual equivalence to other binary relations (such as space improvement) and unary predicates (such as memory safety), as well as to languages with non-monotone store.