Matricial ranges, dilations, and unital contractive maps

summary

The gist

Matricial ranges, dilations, and unital contractive maps investigates whether every unital contractive map on the operator system generated by a Jordan block of size 3 is completely positive in…

In short

The paper investigates whether every unital contractive map on the operator system generated by a 3x3 Jordan block is completely positive in specific cases, providing partial information about its matricial range. It explores structural theorems for dilations of operators satisfying certain conditions and examines related properties for normal matrices.

Key concepts

Matricial Range
The matricial range of an operator is the set of all matrices whose numerical radius is at most 1/2. For a normal operator, this range is described as the C*-convex hull of its spectrum, and for self-adjoint operators, it relates to intersections of closed half-planes.
Structure Theorems for Dilations
The study develops a theorem concerning dilations of an operator B satisfying BB* + B*B = I. The goal is to determine if every such B in Mn satisfying BB* + B*B <= In admits a dilation Be where BeBe* + Be*Be = I. This investigation shows this holds for n=2 but fails for n>=3.
Completely Positive Map
A unital contractive map is completely positive if it satisfies a specific contractive condition: ||alphaX + βX*|| <= p|α|2+|β|2. The paper establishes an equivalence between this contractive condition and complete positivity for maps from OS(J3) to OS(X) when 0 is not in the closure of W(X).
Jordan Block J3
The Jordan block J3 is a specific type of 3x3 matrix used as an operator system. The paper analyzes unital contractive maps on the operator system generated by this block, investigating when they are completely positive and characterizing its matricial range.

Terminology used across episodes

This episode discusses

The paper

Matricial ranges, dilations, and unital contractive maps · Read on arXiv

Pankaj Dey, Atanu Dhang, Mithun Mukherjee

Indian Institute of Technology Bombay · Indian Association for the Cultivation of Science

DOI: 10.4153/S0008414X26102636

Transcript

Introduction to the show: ident: Mathematics. The day in mathematics: new theorems, new proofs, and what they open.

Ada: I'm Ada.

Emmy: And I'm Emmy, here with Kolya.

Ada: We're looking at the paper: "Matricial ranges, dilations, and unital contractive maps".

Emmy: Here's the gist of the paper. Matricial ranges, dilations, and unital contractive maps investigates whether every unital contractive map on the operator system generated by a Jordan block of size 3 is completely positive in certain cases, providing a partial description of its matricial range The paper explores the characterization of these ranges through structural theorems for dilations and examines related properties for normal matrices Matricial Range Characterization The matricial range of an operator is defined as the set of all matrices with numerical radius at most 1/2 Arveson introduced this notion in his seminal article called Subalgebras of C∗-algebras II (Acta Math, 128, 1972) For a normal operator T, the matricial range Wn(T) is described as the C∗-convex hull of its spectrum The description of the matricial range of an operator can be described as intersection of closed half-planes Specifically, for a self-adjoint operator T, Wn(T) = n X ∈ Mn: W(X) ⊆ W(T) o Furthermore, the matricial range of J2 is characterized as the set of all matrices whose numerical radius is at most 1/2 Structure Theorems for Dilations The investigation develops a structure theorem for a dilation of an operator B satisfying BB∗+B∗B = I The authors investigate whether every B ∈ Mn satisfying BB∗+B∗B ≤ In admits a dilation Be for which BeBe∗+Be∗Be = I This investigation shows that this has an affirmative answer for n = 2, but fails to hold in general for n ≥ 3 A structure theorem is provided for a dilation of B ∈ Mn satisfying BB∗ + B∗B = In Contractive Map Properties on J3 The paper proves that every unital contractive map on the operator system generated by J3 is 2-positive Next, it shows that every unital contractive map on the operator system generated by J3 is completely positive in certain cases This leads to a partial description of the matricial range of J3 The equivalence between a unital self-adjoint preserving linear map being completely positive and satisfying the contractive condition αX + βX∗≤ pα2+β2 is established for maps from OS(J3) to OS(X) when 0 ∈/ W(X)◦ Properties of Normal Matrices Dilations The study extends to 4 × 4 normal matrices, showing that studying unital contractive maps on the operator system generated by such matrices is equivalent to studying a unital contractive map on the operator system generated by T = diag(λ, −1, i, −i) where R(λ) ≥ 0 The authors prove that every unital contractive map on the operator system generated by T = diag(1, −1, i, −i) is completely positive Extreme Points of the Convex Set The analysis of the set S = (B1, B2): B1, B2 ∈ (Mn)s.a., B2 1 + B2 2 ≤ In reveals a scheme for determining whether (B1, B2) ∈ E(S) The extreme points of S are characterized by Theorem 4.12, which depends on the rank of X1X2 This characterization shows that for n=3, the extreme points are the union of E1(S) and E2(S) up to unitary similarity Matricial Range Bounds The paper establishes that Wn(J3) is contained in the set of matrices X ∈ Mn: XX∗ + X∗X ≤ In This follows from Corollary 6.9, which shows that for any Y ∈ Wn(J3), there exists a UCP map phi such that αY +βY∗≤ pα2+β2 Conversely, the set of matrices with trace zero and Frobenius norm less than or equal to 1 is contained in W2(J3) The paper concludes by noting that for a normal matrix T obtained by an invertible affine transformation to N = diag(1, −1, i, −i), every unital contractive map from OS(T) to B(H) is completely positive This implies that the set of all unital completely positive maps coincides with the set of all unital contractive maps on the operator system generated by such a normal matrix The authors leave Question 7.5 unanswered regarding whether every unital contractive map from OS(N) to B(H) is completely positive Remarks and Examples The paper provides examples illustrating both contractive and non-contractive maps on operator systems generated by normal matrices Example 8.6 demonstrates a 2-positive map on the operator system generated by a 3 × 3 matrix which is not completely positive The matricial range of Jn is invariant under transposition and rotation Final Conjecture The paper concludes with Conjecture 6.11, which states that every unital contractive map on the operator system generated by J3 admits a positive extension to M3 This conjecture remains open for further investigation in the future The authors note that their MATLAB/Mathematica computations confirm that the second inclusion in Corollary 6.9 is an equality References

1: Ando, T.

Ada: Structure of operators with numerical radius one.

Emmy: Acta Sci.

Ada: Math.

Title and authors: Ada: So we're looking at this paper called "Matricial ranges, dilations, and unital contractive maps" by Dey, Dhang, and Mukherjee. It seems to be tackling a long-standing question about characterizing the matricial range for Jordan blocks of size three <ref:2509.18027#pg1>.

Emmy: Exactly. They are investigating whether every unital contractive map on the operator system generated by this specific Jordan block is completely positive in certain scenarios, which gives us a partial description of that range we're interested in.

Kolya: So, what's the main thrust here? Is it just about characterizing these maps, or are they using this to figure out something bigger about the structure of these operator systems?

Ada: Well, the paper starts by showing that every unital contractive map on the operator system generated by a Jordan block of size three is actually two-positive <ref:2509.18027#pg1,that every unital contractive map on the operator system generated by>. Then, they move on to showing it's completely positive in some specific cases, which helps them describe that matricial range.

Emmy: And to get there, they develop a structure theorem for a dilation of an operator B satisfying the condition BB star plus B star B equals the identity matrix. This is key for seeing how these maps behave under dilation.

Kolya: So, regarding that structure theorem, what's the catch? Does it work universally for all operators satisfying that initial inequality, or does it break down when we go to larger sizes?

Ada: The paper shows that this structure theorem works affirmatively for the size of two, but then it fails for any size of three or greater. This failure is important because it tells us something about the complexity when you move beyond those smaller cases.

Emmy: That failure is a big deal structurally, because it means we can't rely on that dilation structure to solve everything for larger Jordan blocks.

Kolya: Okay, so if we look at the specific results they get for size three, what's the actual implication for us who are just listening and not deep in the operator theory?

Ada: The main implication is that they establish a partial description of what the matricial range of a Jordan block of size three looks like <ref:2509.18027#pg1>. It doesn't give a complete picture, but it gives us constraints based on these contractive maps.

Title and authors: Emmy: And they connect this to another idea: the equivalence between a unital self-adjoint preserving linear map being completely positive and satisfying that specific contractive condition with alpha, beta, and p.

Kolya: That sounds like a strong link. Does this mean if we see that map behaving contractively in that way, we can immediately assume it's completely positive under those conditions?

Ada: Not exactly. The equivalence holds only when zero is not in the closure of the numerical range of X, which is a specific condition on the operator X involved.

Emmy: And they show that for four by four normal matrices, studying contractive maps is equivalent to studying a simpler operator T with eigenvalues one negative one i and negative i.

Kolya: So, when they look at that T with those eigenvalues, what do we actually get? Does it simplify the problem enough to get a definitive answer on complete positivity?

Ada: They prove that every unital contractive map on the system generated by that specific four by four matrix T is completely positive. This result then extends to any normal matrix you can get from an invertible affine transformation to that T.

Emmy: That extension seems very powerful, showing how a specific structure on the eigenvalues dictates the behavior of all unital contractive maps for those systems.

Kolya: But what about the limits? The paper does mention some examples of maps that are two-positive but not completely positive on a three by three matrix <ref:2509.18027#pg1>. How does that fit into this whole picture?

Ada: That example is used to show that even when we have contractive maps on the system generated by a three by three matrix, they aren't always completely positive. It’s a counterexample showing the limitation of just being two-positive.

Emmy: So, it highlights that the condition for complete positivity isn't automatically met just because the map is contractive, even in these small cases.

Kolya: And regarding those extreme points of the set S with B1 and B2, how does characterizing those points help us understand which maps are completely positive?

Ada: They characterize the extreme points using Theorem four point one two, and for size three, they find that these extreme points are just the union of two specific sets up to unitary similarity <ref:2509.18027#pg2>.

Title and authors: Emmy: That structural characterization helps pin down exactly what kind of structure we need in the operator system to guarantee complete positivity for those maps.

Kolya: So, looking at the final results on normal matrices, how certain are we that this equivalence holds for all those affine transformations? Is there a loophole we should watch out for?

Ada: The paper states that it holds because they've shown the equivalence is established through their work with T = diag(one negative one, i, negative i), so for any normal matrix you can get from that via an invertible affine transformation to T, the map is completely positive <ref:2509.18027#pg1>.

Emmy: That gives a very concrete handle on when we can assert complete positivity for these maps.

Kolya: Before we wrap up this discussion on "Matricial ranges, dilations, and unital contractive maps," I want to ask what the biggest open question is here? What's left unanswered?

Ada: They leave Question seven point five unanswered regarding whether every unital contractive map from the operator system generated by that matrix T with those eigenvalues is completely positive.

Emmy: So they show a lot about when it works, but they don't solve the general case for that specific four by four structure.

Kolya: That makes sense. It sets up the next line of inquiry for people who want to push this further, moving from what we know to what we still need to prove.

Ada: Right, so in summary, this paper gives us a partial description of the matricial range of a three by three Jordan block and shows that contractive maps on certain normal matrices are completely positive, but it leaves some specific cases open for deeper investigation.

Emmy: It's a solid piece of work connecting dilations, contractive maps, and complete positivity in this area.

Kolya: I think the connection between the structure theorem for dilations and the results on size two versus size three is really what gives this paper its real substance for us.

Ada: Definitely. It shows that these structural theorems are powerful tools even when they fail to give a complete picture, telling us exactly where those pictures break down.

Emmy: We'll leave it there with the full details on "Matricial ranges, dilations, and unital contractive maps."

The paper's summary: Ada: So we've seen how they looked at these maps on Jordan blocks of size three, and now we need to talk about what this paper actually says is going on with that whole picture.

Emmy: It boils down to them showing a partial map of the matricial range for those three by three blocks, linking it directly to whether a unital contractive map is completely positive under certain conditions.

Ada: Exactly. They’re essentially mapping out the boundary of what's possible for these operator systems based on how well the maps behave contractively.

Emmy: They do this by using two main tools—a structure theorem for dilations and characterizing those extreme points of a set involving two matrices, B1 and B2.

Ada: That structure theorem part is interesting because it works for size two, but it breaks down when you try to apply it to size three or larger blocks. That’s a huge piece of information about the complexity involved in these operator systems.

Emmy: And those extreme points, they characterize them using a condition on the rank of some matrices, Xone and Xtwo which helps them pin down exactly where things are going wrong or right structurally for size three.

Ada: So what does this mean for someone who just listens to the show? It means we're getting a rough sketch of the shape of this matricial range using these maps as a guide.

Emmy: Right, it gives us some concrete constraints on that range, which is important because it connects abstract operator theory to something more visual.

Ada: It’s like saying if you see a certain kind of behavior in the map—a contractive one—then you can get hints about where that matrix falls within the allowed region.

Emmy: And they establish this link between being completely positive and satisfying that specific contraction condition with alpha and beta, which only happens when zero isn't in the closure of the numerical range of X.

Ada: That’s a tight constraint. It means complete positivity isn't just about the map being contractive; it depends on where you are in terms of those ranges.

Emmy: And they also found that for normal matrices that can be transformed into a specific four by four matrix with eigenvalues one, negative one, i, and negative i—that unital contractive map is completely positive.

Ada: That’s the part where they get a definitive win; if you have that specific structure on the matrix you started with, the map is guaranteed to be completely positive.

Emmy: But they also flag some things that aren't working, like examples of maps on three by three matrices that are two-positive but definitely not completely positive.

Ada: So it’s not a universal rule; there are still exceptions even in these small cases, which is where the nuance lies.

Emmy: This paper gives us the framework for understanding those nuances, showing how dilations and contractive conditions interact to define these ranges.

The paper's improvements: Ada: So we're looking at how the authors suggested ways to make this paper even better, and they focus on tightening up those structural theorems we talked about earlier.

Emmy: They suggest refining the way they handle those dilations of operators satisfying that identity matrix condition, BB star plus B star B equals I.

Ada: It’s like if you have a map for a road but you can make the lanes wider or narrower depending on where you are, they’re suggesting ways to make that lane structure more precise.

Emmy: They also look at making the characterization of those extreme points of the set involving B1 and B2 even more specific by using some theorems about rank and similarity.

Ada: That would help us figure out exactly what kind of structural arrangement we need in the operator system to get a complete picture for size three blocks.

Emmy: And they explore how that link between complete positivity and the contraction condition works when zero is not in the closure of those numerical ranges, which is a big restriction on when that equivalence holds.

Ada: That restriction tells us exactly what kind of operator we need to be looking at for the math to work cleanly.

Emmy: Plus, they seem interested in seeing if they can make the connection between contractive maps and complete positivity even tighter by exploring the bounds on those maps.

Ada: They want to know if there are better ways to define those bounds so we can see exactly where a map fails to be completely positive.

Emmy: And they’re looking into that open question about whether every unital contractive map on that specific four by four matrix system is actually completely positive.

Ada: That's the next big thing for them, pushing past the known results on size three and looking at these larger, more structured normal matrices.

Emmy: It seems like they’re trying to build a whole toolkit here to move from partial descriptions to a more rigorous characterization of those ranges.

Ada: So they are aiming for a clearer rulebook on when you can predict complete positivity just by observing contractive behavior.

Conclusion: Ada: So we're wrapping up this look at "Matricial ranges, dilations, and unital contractive maps." Basically, they've given us a solid set of constraints on what those matricial ranges look like for these three by three blocks.

Emmy: It’s a piece of work that ties together operator theory with geometry—showing how dilation structure dictates the behavior of these maps.

Ada: They proved that for certain normal matrices, if you have a unital contractive map, it is completely positive, and they even found the precise conditions under which that equivalence holds.

Emmy: It’s important because it shows us a pathway to test for complete positivity just by checking if the map satisfies those specific contraction bounds.

Kolya: From a probabilistic standpoint, what this means is that we have some very specific tail behavior governed by these maps, and we can use that structure to set bounds on how far the map can stray from being completely positive.

Ada: Right, so even if the map isn't perfectly completely positive, knowing it's two-positive gives us a starting point for where it’s headed.

Emmy: And they leave that question about whether every unital contractive map on that specific four by four matrix system is completely positive hanging in the air.

Ada: It leaves room for more structural exploration, especially when we look at how those extreme points of the set S behave under different conditions.

Emmy: That opens up a lot of avenues for future work in understanding these operator systems more deeply.

More episodes

← Home