Proofs Theorem 1

Theorem 1: Any sequence of infrakernels which fulfill the niceness conditions have the infinite semidirect product being well-defined and fulfilling the niceness conditions.

As a recap, the niceness conditions are:

1: Lower-semicontinuity for inputs: For all continuous bounded , is a lower-semicontinuous function.

2: 1-Lipschitzness for functions: For all , is 1-Lipschitz.

3: Compact-shared compact-almost-support: For all compact sets and , there is a compact set which is an -almost-support for all the inframeasures where .

4: Constants increase: For all and constant functions , .

5: 1-normalization (only for the type signature): For all ,

Now, is defined as: Fixing an arbitrary sequence of compact sets ,

Our task is to show that is a well-defined infrakernel which fulfills these five properties.

We will be reusing the proofs from “Less Basic Inframeasure Theory” (henceforth abbreviated as LBIT) when possible, and commenting on the differences when applicable.

The proof sketch is:

Part 1: Show that the functions you’re feeding into those infrakernels are guaranteed to be continuous.

Part 2: Show that all the are infrakernels which fulfill the niceness conditions, via induction.

Part 3: Show that if a function only depends on the first n coordinates of the input, then the values monotonically increase as m does.

Part 4: Give a general procedure for taking a compact subset of the space and making a compact subset of the space with nice properties related to compact almost-support, that preserves its nice properties when projected down to any finite stage.

Almost-Monotone Lemma: This is a sufficiently useful utility tool in other theorems to be broken out into its own result. It says you can take any and and compact set and pick a sequence of compact sets for the defining sequence of the infinite semidirect product such that the defining sequence for all the (with ) is -almost-monotone.

Part 5: Use parts 2, 3, and 4 to show that for any two different sequences, the defining sequences for the infinite infrakernel will limit to each each other (so if one converges, all do, if one diverges, all do, and if one wiggles around forever, all do), and then apply the Almost-Monotone Lemma to rule out the “wiggle around forever” case. Thus, the limit to define exists, and doesn’t depend on the choice of the sequence of compact sets. This solves well-definedness.

Part 6: Using parts 2 and 5, clean up the basic inframeasure conditions for (monotonicity and concavity), and show 1-Lipschitzness, increase of constants, and 1-normalization for while we’re at it. That’s 35 niceness conditions down, and all inframeasure conditions except for compact almost-support.

Part 7: Use our trick from Part 4 and our freedom of picking our compact set sequence from Part 5 to show the compact-shared compact-almost-support property for . This also means that individual outputs of the kernel fulfill CAS, which verifies the last condition we need to conclude that is always an inframeasure. Only lower-semicontinuity of the kernel remains to be verified.

Part 8: We use the Almost-Monotone Lemma to wrap up the last condition, the lower-semicontinuity of the infinite infrakernel.

We often use the usual “start with thing we’re trying to prove, and keep reducing it to a simpler problem” technique from the last time this proof path was used.

T1.1 Our desired result is whether the function
is continuous. This was already shown in “Less Basic Inframeasure Theory” (LBIT).

T1.2 The desired result is “if all the have the five niceness properties, then all the are infrakernels with the five niceness properties as well”.

The proof sketch for this is that this splits into 6 properties we need to show induct up. Monotonicity (phase 1), concavity (phase 2), 1-Lipschitzness (phase 3), increasing constants (phase 4), shared compact-almost-support (phase 5), and lower-semicontinuity (phase 6), which splits into two parts, one where we show a particular function is continuous, and then show lower-semicontinuity in general. Then, we finish up by arguing that this gets everything we need.

For the base case, because and we’re assuming all the have the niceness properties and are infrakernels, trivially fulfills them.

Time for the induction step. Remember that

T1.2.1 Proof of monotonicity: Assume . Then

And the same goes for so if we could prove


we’d be successful. Now, by monotonicity for (by induction assumption), we’d have that result if


Letting be arbitrary, by applying monotonicity for , we’d have that result if

Which is true since . And so we’re done, montonicity inducts up.

T1.2.2 Proof of concavity:

and then, by concavity for all , and monotonicity for (by induction assumption), we have


and then, by concavity for (by induction assumption), we have


and then this packs up as

and concavity inducts on up.

T1.2.3 Proof of 1-Lipschitzness: we have


And by 1-Lipschitzness for (by induction assumption), we have


And by 1-Lipschitzness for , we have

And 1-Lipschitzness inducts on up.

T1.2.4 Proof of increasing constants (and 1-normalization): We have

And then by increasing constants for , and monotonicity for (induction assumption), we have

Now just apply increasing constants for (induction assumption) to get

For 1-normalization in the type signature, we can have , and then instead of an inequality, we have an equality, and the same proof path works, just using 1-normalization instead of increasing constants. They both induct up.

T1.2.5 Proof of the compact-shared CAS property: Our task is to take an arbitrary compact set , and find a compact set where two functions identical on the set only differ in expectation by according to all with .

Let , and define the set as

where the first compact set is a shared -almost-support for the where (exists by compact-shared CAS for , an induction assumption) and the second compact set is a shared -almost-support for the where is in (exists by compact-shared CAS for , an assumed niceness condition). Let be continuous and identical on . Then, we have


Now, we apply Lemma 2 from LBIT, where we can upper-bound this quantity by “Lipschitz constant of the inframeasure times how much the functions differ on the almost-support + level of almost-support times how much the functions differ in general”. We use as our compact set for the inner function. It was defined as a -almost support for all with , and is 1-Lipschitz (induction assumption). So our upper bound is:




Focusing on that second chunk, we can apply 1-Lipschitzness of to yield



For the first chunk, we can apply Lemma 2 from LBIT again, with as our compact set, which is a -almost support for all with , and is 1-Lipschitz. So our upper bound is:



Now, since and are identical on that first term just vanishes, yielding


We upper-bound this by


Which then clearly just reduces to

And we’re done, the compact-shared CAS property inducts up.

T1.2.6.1 Proof of lower-semicontinuity: We’ll start by taking an extended detour to show that the function

is lower-semicontinuous. Fix a convergent sequence of inputs, which limit to . First up, the set

is compact in since it converges and we have the limit point added in. By the compact-shared CAS property for , we can take any and make a compact set which is a shared -almost-support for all the inframeasures. For any m, we can then apply the Lemma 2 (from LBIT) decomposition with the -almost-support for all the inframeasures (and 1-Lipschitzness of ) to get the inequality


That second term can be upper-bounded by , so doing that, we have

Now, the set is a compact set, so any continuous bounded function is uniformly continuous on it. For all , there is some where two points only apart within this set have only differing by on them. Since limits to , for sufficiently large m, those two points will be within of each either, so eventually

will be or less for late enough m. So, for late enough m, we have

Since can be arbitrary, we have that the sequences

and

limit to each other. Therefore,

And then, by lower-semicontinuity for , we have

and so, lower-semicontinuity of the function

is shown. Time to return to the usual induction proof.

T1.2.6.2 For lower-semicontinuity in inputs, we first unpack the semidirect product. Fix a sequence which limits to .

Since we showed that the function

is lower-semicontinuous, we can apply the monotone convergence theorem for inframeasures and rewrite as

For any individual , call it , we have

We now run through the same proof path as earlier again. The set is compact in since it converges and we have the limit point added in. By the compact-shared CAS property for (induction assumption), we can take any and make a compact set which is a shared -almost-support for all the inframeasures. For any m, we can then apply the Lemma 2 (from LBIT) decomposition with the -almost-support for all the inframeasures (and 1-Lipschitzness of by induction assumption) to get the inequality


That second term can be upper-bounded by , so doing that, we have

Now, the set is a compact set, so any continuous bounded function f is uniformly continuous on it. For all , there is some where two points only apart within this set have only differing by on them. Since limits to , for sufficiently large m, those two points will be within of each either, so eventually

will be or less for late enough m. So, for late enough m, we have

Since can be arbitrary, we have that the sequences

and

limit to each other. Taking a break to recap, we had

and for any fulfilling those bounding conditions for the sup, we have

and from our fresh new result, we then have

and, by lower-semicontinuity for (induction assumption), we have

Since was arbitrary below

this means that

Now, we can put our lower-semicontinuous upper bound back in to get

and we’re done, induction applies for lower-semicontinuity.

All our induction steps have been shown. First, for showing that all the are inframeasures, we have monotonicity and concavity. Lipschitzness is taken care of by our 1-Lipschitzness induction. Compact almost support is taken care of by our compact-shared CAS induction for the , because a single point is compact. And weak normalization (0 must map to 0 or more) is taken care of by our Constants Increase induction. So all the are inframeasures.

For the 5 niceness conditions, we proved all of lower-semicontinuity, 1-Lipschitzness, compact-shared CAS, increasing constants, and 1-normalization by induction. So we can now assume that all are nice.

T1.3 We must show that if we go far enough out in the , the value assigned to functions which only depend on finitely many inputs starts monotonically increasing. The result that we’d like to show at this point is:

Fix an arbitrary and . Then, we can just go:

Since , that function at the end is a constant (doesn’t depend on its input), so by Increasing Constants for and Monotonicity for , we have

and we’re done.

T1.4 Our desired result here is:



This says that if you pick any compact subset of , you can make a compact subset of where the projection of it to coordinates 1 through acts as a compact -almost-support for all the infradistributions when lies in your compact set .

The proof of this is word-for-word identical to the proof of this from LBIT, nothing changes. To recap the construction, fix some and . Now, we recursively build up compact subsets of all the via the following inductive process.

is an arbitrary compact shared -almost-support for where is in , these sets exist by compact-shared CAS for all the .

We define the abbreviations

and

This is the product of all the compact sets, and is compact. Notice the dependence of these on the starting compact set and . Notice that the projection of to coordinates 1 through n is exactly .

Almost-Monotone Lemma: If is a sequence of infrakernels fulfilling the niceness conditions, then



This says that for any compact subset of and you name, I can find a “good” sequence of compact sets where the defining sequence for all the quantities is “almost monotone”. Every time the sequence hits a certain value (at time n), it will never again be able to dip below of that value as the sequence progresses. This proof progresses in 5 phases. The first phase is just doing some rewrites of our proof goal, the second, third, and fourth are showing setup results on functions undershooting other functions, and certain pairs of quantities being similar. The last phase is wrapping up the actual result.

L1.1 So, for the rewrites, let be an arbitrary compact subset of , and be arbitrary. Our proof goal is now



Let your sequence of compact sets be the sequence recursively defined from and . Then our proof goal becomes



Let , and be arbitrary. Remember is in . Then our proof goal becomes


First, let’s define some abbreviations for later. The function

will be abbreviated as , and the function

will be abbreviated as . Both of these functions are continuous, by section 1. We’ll need three key results before beginning.

L1.2 Our first desired result is:

Unpacking this proof target, it is


If , this result is trivial, we have equality. So now we’ll assume . Observe that it is now possible to factorize as , therefore,

turns into

which is just

Packing up on the left side of our proof target, our proof target is now


And then, since we can factorize as , our proof target is now


Which is obviously true. So, we have our first result of

L1.3 Time for our second desired result. Our second desired result is that

For this, since on the set , as we showed in our first desired result, equals on the set .

Further, is an -almost-support for all the with , as our is. So, in particular, it’s an -almost-support. Accordingly, we have

Now, for any input, either is higher (and so the distance between the functions at that point is 0), or is higher, and so the distance between the functions at that point is upper-bounded by . So, we have

At this point, we recall that both and were just , but taking worst-case inputs past a certain point, so a trivial bound on this quantity is . So, our net result is

L1.4 And it’s time to move onto our third setup result. Our desired third setup result is that

For this, since on the set , as we showed in our first desired result, equals on the set .

Further, is an -almost-support for all the with , as our is. So, in particular, it’s an -almost-support. Accordingly, we have

Now, for any input, either is lower (and so the distance between the functions at that point is 0), or is lower, and so the distance between the functions at that point is upper-bounded by . So, we have

At this point, we recall that both and were just , but taking worst-case inputs past a certain point, so a trivial bound on this quantity is . So, our net result is

Now that everything’s in place, we can start working on our actual result.

L1.5 Remember, our overall proof target at this stage is


We can reshuffle this proof target a little to get the equivalent statement


So let’s get cracking on that inequality.

First, we abbreviate it as

Then, we apply our second sublemma we proved to get

And undo the absolute value to get

and cancel out to get

and then, by monotonicity we have

and apply our third sublemma we proved to get

And undo the absolute value

And cancel

and at this point, we can unpack the inner function to get

And then, we apply our Part 3 result that for functions which only depend on finitely many inputs, is monotonically increasing, to get

and we’re done with the Almost-Monotone Lemma!

T1.5 Our aim here is to show that no matter what sequence of compact sets you have, they all limit to the same result. In order to do this, we’ll have to show the result (letting being your first sequence of nonempty compact sets to attempt to define the limit and being your second sequence of nonempty compact sets, and abbreviating as the product of the from to ) that,



In words, this is saying that for any two sequences of compact sets, there exists some threshold where if you pick any value of the defining sequence for associated with using as your compact sets, and the sequence associated with using as your compact sets (as long as they’re both past some shared threshold), they’ll be close. So all choices of compact set must limit to each other, because can be arbitrary.

Fix our (input value, closeness parameter, two sequences of compact sets, function), and now we’ll find our n to make


true. Do it in the following way. Use as your compact seed set to invoke the technique in part 4 to construct your sequence , and then consider the set:

A finite union of compact sets is compact, and the product of compact sets is compact. If we restrict to this set, it’s uniformly continuous. In particular, using the standard product metric (which metrizes the product topology), defined as:

We can conclude that two sequences which agree up till time n must be, at most, distance apart. Since restricted to our compact set of interest is uniformly continuous, there is some difference in inputs that only leads to an difference in values. Now we can define our n as .

Now that we’ve picked our n, let m be arbitrary. Our goal is to now prove:


First-up, we use that is an -almost support for , and 1-Lipschitzness, to apply Lemma 2 from LBIT to break up


as




We can upper-bound the second term with just , as the difference in values can’t be any higher than the gap between the maximum value of and the minimum value of . As for the first term, note that if and and (the latter two are the inf terms for a particular ) then we have

And same for . And these two points agree on , and n is large enough that the function itself varies by only for any two points in that set which share the same prefix up to length n (and fulfills that) So, we can break down the upper bound as

And so we’re done here and have our result.

Moving on, the statement



is the same as



So all the defining sequences for limit to each other. We’ve got three possible cases. One is divergence, that liminf of the defining sequences goes to infinity. This is fine, the limit exists. One is convergence, that liminf and limsup are finite and equal, ie, the limit exists. This is fine. And the third case, that liminf and limsup aren’t equal, must be ruled out. We’ll invoke the Almost-Monotone Lemma, that



Specializing to the compact set consisting of a single arbitrary point , and recalling that the defining sequence for that was the sequence of almost-supports, and an arbitrary function , the lemma turns into



Fix an arbitrary and now. Just by definitions, we have

This sequence is monotonically increasing, so a lower bound on this is

and then, using and in the Almost-Monotone Lemma, we have

Since was an arbitrary number, we have

and then we can lower-bound this by the limsup, yielding

Since the limsup is always equal or greater than the liminf, this implies that the limsup and liminf can be at most apart for this sequence. And since all the defining sequences limit to each other, they all have the same limsup and liminf. Since was arbitrary, we get that all the defining sequences limit to each other, and either converge to a fixed number, or diverge to infinity. The infinite semidirect product is now well-defined, and we can use whichever sequence of compact sets is the most convenient.

T1.6 Showing the following infradistribution properties for the infinite semidirect product: monotonicity and concavity. Also, the 1-Lipschitz property and 1-normalization property and Constants Increase property, for 45 inframeasure properties, and 35 niceness properties.

For monotonicity, concavity, and 1-Lipschitzness, the exact proof from LBIT works.

For constants increasing, observe that regardless of n,

Then, this bound transfers to the limit, so

The same proof applies to 1-normalization, but with equality signs, if we plug in 1, and the limit of the all-1 sequence is always 1.

T1.7 Showing compact-shared compact-almost-support for . Again, the exact proof from LBIT works here. We’re up to 55 inframeasure properties, and 45 niceness properties for the infinite semidirect product.

T1.8 Our task is to show lower-semicontinuity for . This will take some cunning. At the start, fix an , we’ll see why later. Our task is to show that

For some arbitrary sequence of which limits to . What we can do is first unpack.

Now, the sequence of sets we picked was important. It means we can now apply the Almost-Monotone Lemma. Regardless of m, the defining sequence for is almost monotone now. We can fix a particular (arbitrary) to get a lower bound of

And then, by lower-semicontinuity for , which was already established by induction, we get

But, since this holds for all , we can take a limit and get

But since can be arbitrarily small, we have

and thus, lower-semicontinuity. And we’re finally done!!

No comments.