[00:00:01] Speaker 03:
Probably four argued cases this morning, and counsel ready to proceed with the first?

[00:00:09] Speaker 03:
Is there a Keeson or a Kesson?

[00:00:11] Speaker 03:
Keeson.

[00:00:12] Speaker 03:
Keeson.

[00:00:13] Speaker 03:
Please proceed.

[00:00:15] Speaker 03:
You're reserving three, right?

[00:00:16] Speaker 03:
Correct.

[00:00:17] Speaker 03:
Yeah.

[00:00:18] Speaker 02:
Thank you, Your Honors.

[00:00:19] Speaker 02:
Good morning.

[00:00:20] Speaker 02:
May it please the court?

[00:00:21] Speaker 02:
I'm Jay Keeson for the Appellant Invespic.

[00:00:24] Speaker 02:
The key aspect of the invention at issue here is that it allows the user to model

[00:00:30] Speaker 02:
multiple investments in a portfolio, while analyzing these investments in a different market condition.

[00:00:38] Speaker 02:
And by doing so, by sampling these investments at the same time, you maintain the temporal correlation.

[00:00:45] Speaker 02:
In this case, the board committed multiple legal errors, all of which flow from its failures.

[00:00:49] Speaker 03:
Let me ask you a couple of housekeeping questions, OK?

[00:00:57] Speaker 03:
Where in the record

[00:00:59] Speaker 03:
Do you offer the board claim constructions for the terms bias parameter and resampling process?

[00:01:08] Speaker 02:
Sure, Your Honor.

[00:01:10] Speaker 02:
For bias parameter, support in the record below can be found on pages 1245 to 1247 and from pages 1372 to 1375.

[00:01:28] Speaker 02:
For resampling, there are a number of places where our definition is supported, but the best ones are at 1256, 1266, 1297, and 1384.

[00:01:47] Speaker 02:
So that's with respect to both bias parameter and resampling.

[00:01:51] Speaker 02:
Does that answer your question?

[00:01:52] Speaker 03:
Well, I don't know.

[00:01:53] Speaker 03:
You say there's support for them.

[00:01:55] Speaker 03:
You offered a definition there.

[00:01:57] Speaker 02:
Correct.

[00:01:57] Speaker 03:
OK.

[00:01:58] Speaker 03:
Correct.

[00:01:58] Speaker 03:
I'll look at it.

[00:01:59] Speaker 03:
Correct, Your Honor.

[00:02:01] Speaker 01:
Can I ask you a question about the bias parameter?

[00:02:06] Speaker 01:
And I'm probably going to be inarticulate or incorrect even in the formulation of the question, so I'm going to hope that you'll help me.

[00:02:17] Speaker 01:
On the bias parameter, your key textual point, as I understand it, is that the bias parameter has to be in

[00:02:27] Speaker 01:
the picking of samples from a preselected pool and that's not done in Sortino.

[00:02:35] Speaker 01:
The board, as I understand it, pointed to without, again as I understand it, without really distinguishing two different things that might be a bias parameter or might be in combination a bias parameter.

[00:02:55] Speaker 01:
One is

[00:02:56] Speaker 01:
the splitting of the overall sample space into seven scenarios, and then after each scenario has been randomly sampled and a particular statistical measure has been created for each one of those, some sort of weighting of the results.

[00:03:15] Speaker 01:
Now, I understand again textually why you might say that that's not

[00:03:21] Speaker 01:
a bias parameter in the selection of the sample.

[00:03:24] Speaker 01:
Now here's my question.

[00:03:28] Speaker 01:
Can you explain to me, in a way I might understand, why as a mathematical matter, using a bias parameter in the selection of the samples from the already defined sample space is in any way different from weighting the measures that you derive from

[00:03:51] Speaker 01:
a random resampling process within each scenario.

[00:03:55] Speaker 02:
So a difference, Your Honor, and Your Honor is exactly correct in characterizing the arguments.

[00:04:05] Speaker 02:
The key issue is that if you were to bootstrap and then you were to combine them, then you lose the temporal correlation between investments.

[00:04:21] Speaker 02:
So what the BARMA patent teaches is it allows you to sample multiple investments, let's say on the same day, and how they relate with respect to each other.

[00:04:34] Speaker 02:
Sortino bootstraps each investment in each scenario.

[00:04:39] Speaker 02:
And when you bootstrap it, you've lost the link between this investment and this other investment, how they relate to each other on, for example, the same day.

[00:04:49] Speaker 02:
So you can think of it this way.

[00:04:51] Speaker 02:
It's sort of like if you imagine you had a bunch of ping pong balls in a lottery basket, if you were to pull a ping pong ball for one day and that had all the values of your investments on that day, then you wouldn't lose the temporal correlation.

[00:05:05] Speaker 02:
So you would know that on this day, this investment is up and this investment is down, and then on a similar down day, you may have a different relation, but you've kept track of it.

[00:05:15] Speaker 02:
If you just sample one scenario,

[00:05:17] Speaker 02:
then you lose it.

[00:05:18] Speaker 02:
Does that answer your question?

[00:05:21] Speaker 01:
Part of what I'm not sure about is whether that answers my question or not, I'm afraid.

[00:05:25] Speaker 05:
Doesn't your answer assume that Sortino was only analyzing one investment at a time?

[00:05:31] Speaker 05:
That is actually he explicitly... But your answer to just now, as I understood it to Judge Toronto, was assuming that that was true.

[00:05:38] Speaker 05:
But then what happens if you're wrong on that?

[00:05:41] Speaker 05:
What happens if we think that Sortino does teach

[00:05:44] Speaker 05:
because it's looking at the S&P 500 and there's a composite of things in the 500, the rationale that was expressed at least in one of the two opinions by the same judge using different rationale.

[00:05:57] Speaker 05:
What happens if, what's your answer to Judge Taranto?

[00:06:01] Speaker 02:
First, Your Honor, we don't have to surmise because Sortino in footnote four explicitly says that, in footnote four of Sortino, he says that they bootstrap a given scenario for a given investment

[00:06:14] Speaker 02:
So there is no way to keep track of multiple investments.

[00:06:18] Speaker 02:
So if you just bootstrap one scenario.

[00:06:20] Speaker 01:
But I guess, I mean, Judge Clevenger quite helpfully clarified my question.

[00:06:26] Speaker 01:
My question was, let's forget about Sortino.

[00:06:30] Speaker 01:
If you do, say, two investments and you're doing resampling on each one, what mathematically

[00:06:41] Speaker 01:
is the difference between the random resampling followed by some sort of linear combination of the second moment or the third moment or the fourth moment or something and biasing the sample selection in the process of choosing your samples.

[00:07:02] Speaker 02:
So, Your Honor, if you were to, say, sample one investment and you bootstrap that investment, you're picking out more samples

[00:07:11] Speaker 02:
But what you're not doing is you're not keeping track of the temporal relation with respect to the other investment.

[00:07:15] Speaker 02:
You're doing each of them separately.

[00:07:17] Speaker 02:
So you might end up a different number of up days and down days, and you might end up a different number of up days and down days.

[00:07:22] Speaker 02:
And they're two separate buckets.

[00:07:24] Speaker 02:
And you haven't kept track of what happens on the same day.

[00:07:27] Speaker 02:
When one goes up, what happens?

[00:07:28] Speaker 02:
It may go up or go down.

[00:07:29] Speaker 02:
And another day, when the market is down, it may behave differently.

[00:07:32] Speaker 02:
So the temporal correlation becomes important.

[00:07:35] Speaker 02:
That's why you keep the temporal correlation, and then you bootstrap.

[00:07:38] Speaker 02:
You don't bootstrap separately.

[00:07:40] Speaker 05:
What happens if there's a disagreement of opinion?

[00:07:43] Speaker 05:
It seems to me that the $64,000 question in both of these cases is what does Sardino teach with regard to one or two and what does Sardino teach with regard to a bias in parameter.

[00:07:56] Speaker 05:
What a reference teaches is a question of fact, right?

[00:08:00] Speaker 05:
So we're sitting here trying to decide what Sardino is telling us and we're having a disagreement.

[00:08:07] Speaker 05:
between rational people who have some skill in the art and understanding what these references teach.

[00:08:12] Speaker 05:
So if substantial evidence is the test, which it is, then why isn't there substantial evidence to support your adversary's point of view in terms of his view of what Sortino teaches?

[00:08:25] Speaker 02:
First, Your Honor, even before we get to the issue of comparing Sortino to the claims, even with respect to, say, bias parameter, the claim

[00:08:37] Speaker 02:
as amended clearly says, and this is a matter of law, the claim says you set a degree of randomness by the user.

[00:08:45] Speaker 05:
The claim says in.

[00:08:46] Speaker 05:
It has to do in.

[00:08:47] Speaker 02:
In sample selection.

[00:08:48] Speaker 02:
In a resampling process.

[00:08:49] Speaker 05:
So does that mean in connection with, in relation to?

[00:08:52] Speaker 05:
Does the word in mean in, in, like there is water in this vessel?

[00:08:59] Speaker 05:
Or does it mean in, meaning in, in relation to, in connection with?

[00:09:04] Speaker 05:
The word has to be defined, right?

[00:09:07] Speaker 05:
And if you look at basically... And if you have two different interpretations that one suggests one interpretation, one suggests the other.

[00:09:16] Speaker 05:
If in means in relation to, then you have to sit down on this issue.

[00:09:22] Speaker 02:
Your Honor, all that we're saying is that it just simply means what the patent teaches us.

[00:09:27] Speaker 02:
And the patent tells us that in-sample selection means that you set the bias parameter when you

[00:09:36] Speaker 02:
pick the samples and that's what we call them.

[00:09:39] Speaker 05:
When Serpino is picking samples, he's picking different months, right?

[00:09:44] Speaker 02:
Or it could be 50%.

[00:09:46] Speaker 05:
No, but he's making some random selections upfront in terms of the months and the years he's picking, right?

[00:09:57] Speaker 05:
Actually, what Serpino does... And why isn't that biased?

[00:09:59] Speaker 02:
Because it certainly affects the outcome.

[00:10:01] Speaker 02:
Answer me.

[00:10:03] Speaker 02:
Why isn't that biased?

[00:10:04] Speaker 02:
Right.

[00:10:05] Speaker 02:
I will, Your Honor.

[00:10:06] Speaker 02:
So Tino does not talk about any bias with respect to the seven market conditions.

[00:10:11] Speaker 02:
What he says is the seven market conditions are exclusive and exhaustive.

[00:10:15] Speaker 02:
And that's what he describes.

[00:10:16] Speaker 05:
He says these are seven market conditions.

[00:10:18] Speaker 05:
I'm talking about the months you pick.

[00:10:20] Speaker 05:
Oh, I'm sorry, Your Honor.

[00:10:22] Speaker 05:
You have to listen carefully because what I'm trying to get at is whether or not a reasonable person, a rational person from a substantial evidence point of view

[00:10:31] Speaker 05:
can look at Sortino and say, yeah, I see there's biasing in there.

[00:10:34] Speaker 05:
And the biasing is in connection with selecting the samples and in connection with applying the process.

[00:10:41] Speaker 02:
Right.

[00:10:41] Speaker 02:
And Your Honor, with respect to the 12 months, there is no biasing.

[00:10:46] Speaker 02:
He picks a scenario, and he samples the investment in that scenario.

[00:10:50] Speaker 02:
There is no teaching in Sortino.

[00:10:54] Speaker 01:
Just to see if we're on the same page, I thought your point about

[00:11:00] Speaker 01:
the scenario selection is that's the definition of the sample space, which the claim expressly distinguishes from in the selection of the sample, not in the definition of the sample space.

[00:11:12] Speaker 01:
Correct.

[00:11:13] Speaker 01:
Correct, Your Honor.

[00:11:14] Speaker 01:
Correct, Your Honor.

[00:11:15] Speaker 01:
Absolutely.

[00:11:17] Speaker 01:
But then it takes the results of the random sampling within each of seven different sample spaces, and it does some weighting.

[00:11:26] Speaker 01:
And I'm trying to understand why that

[00:11:28] Speaker 01:
doesn't amount to biasing in the selection of the sample.

[00:11:34] Speaker 02:
And it simply doesn't because when you do it the way the 291 pattern does it, you can pick a down day or you can pick an up day and then you look at all the values of the investment that you have in your portfolio and you keep them together.

[00:11:53] Speaker 02:
You don't bootstrap them separately.

[00:11:55] Speaker 02:
When you bootstrap them separately, you've lost critical information.

[00:11:59] Speaker 02:
You've lost information about how these two investments relate to each other.

[00:12:03] Speaker 02:
That's the critical insight.

[00:12:04] Speaker 02:
And that is what the 291 Packing teaches as maintaining temporal correlations to perform a better analysis, more accurate analysis.

[00:12:12] Speaker 02:
So that way you don't

[00:12:13] Speaker 02:
underestimate risk and overestimate return.

[00:12:16] Speaker 02:
Because this temporal relation, understanding how they relate to each other, is the key.

[00:12:20] Speaker 02:
And that's what people don't care about.

[00:12:21] Speaker 01:
What are the best declarations or factual materials or evidence in the record to try to get at what we've just been talking about, about why the scenario plus waiting doesn't amount to

[00:12:39] Speaker 01:
bias parameter that affects the degree of randomness in the selection of samples.

[00:12:44] Speaker 01:
So let me give you three examples.

[00:12:47] Speaker 02:
Let me give you three quick examples, Your Honor.

[00:12:51] Speaker 02:
So Dr. Savage, who has a declaration starting on page 571, particularly on page 591, he discusses this ping pong example that I just gave.

[00:13:05] Speaker 02:
Then you have the testimony of Dr. Herzog.

[00:13:08] Speaker 02:
Also, at page 310 and 311.

[00:13:11] Speaker 02:
And finally, you have Dr. Sortino himself.

[00:13:14] Speaker 02:
Dr. Sortino himself testified that the 291 system is completely different.

[00:13:20] Speaker 02:
Those are his exact words on page 291 of the record.

[00:13:23] Speaker 02:
So Dr. Sortino himself said that.

[00:13:26] Speaker 02:
So those are three examples, Your Honor.

[00:13:31] Speaker 02:
I'm going to reserve the rest.

[00:13:39] Speaker 00:
Good morning, Your Honors, and may it please the Court.

[00:13:43] Speaker 00:
I do want to address some of these issues with Sortino and hopefully clarify for the Court the teaching of the reference where the substantial evidence is that supports the Board's findings.

[00:13:53] Speaker 00:
But I first want to address the waiver, because if this Court determines that the patent owner waived the claim construction issues, then we never reach the Sortino issues, because their entire basis for patentability, the entire argument we're making now as to why the things are patentable, relies upon

[00:14:08] Speaker 00:
these narrow claim constructions of the term resampling, of the term bias parameter, of the two or more investments limitation.

[00:14:16] Speaker 01:
I'm not sure what you mean by narrow.

[00:14:19] Speaker 01:
Let's take the bias parameter.

[00:14:23] Speaker 01:
I took it that their point was, I guess, what I tried to describe as my understanding of the point, that what's in Sortino produces certain weightings

[00:14:36] Speaker 01:
produces creation creates different sample spaces but what it absolutely does not do is bias the selection of samples from a particular sample space you've already created and that's different that's did they not make that argument to the board they made that argument to the board let me let me

[00:14:55] Speaker 00:
explain to you, your honor, why the board made the correct decision, why there's substantial evidence to support the board's decision that that is, in fact, disclosed in Sorteem.

[00:15:02] Speaker 00:
So let me start with what's disclosed in the 291 patent, unembodyment that's disclosed in the 291 patent.

[00:15:08] Speaker 00:
Right?

[00:15:08] Speaker 05:
They made that argument, they preserved the right to make that argument here.

[00:15:11] Speaker 00:
But not the claim construction, not the narrow definition that they're urging for the bias parameter, not the language they're trying to read into the bias parameter, which is

[00:15:19] Speaker 00:
in the initial selection of samples.

[00:15:21] Speaker 05:
Why do they have the explication that Judge Tarano just stated, and have him buried in an acclaimed construction?

[00:15:29] Speaker 05:
AKA.

[00:15:30] Speaker 05:
I mean, why does he have to say AKA?

[00:15:32] Speaker 05:
He said, this is my view of what's, you know.

[00:15:37] Speaker 00:
Because the argument that was presented to the board was a distinguishing argument over the prior art.

[00:15:43] Speaker 00:
What they're arguing to you now is,

[00:15:45] Speaker 00:
the board erred by failing to construe the claim in a particular way, by not reading into the claim language or the construction of the bias parameter, that it's an initial selection of samples from the sample space.

[00:15:59] Speaker 00:
That initial selection and that from the sample space, you won't find those words in the arguments that were made to the board.

[00:16:06] Speaker 03:
What I got out of pushing the record to which your opposing counsel cited me and what I got

[00:16:12] Speaker 03:
initially was that they argued against your position, but not that they argued for a position.

[00:16:18] Speaker 00:
I think you're correct about that, Your Honor, that they referred an parenthetical to an argument that we had made that the resampling process is broader than just the sampling of individual monthly samples.

[00:16:30] Speaker 00:
And so they cite to this court an argument that we had made that they had criticized below.

[00:16:34] Speaker 00:
I believe you're correct about that.

[00:16:36] Speaker 00:
Let me explain Sortino versus the 291 patent.

[00:16:39] Speaker 00:
An embodiment in the 291 patent says,

[00:16:42] Speaker 00:
Let's pick a sample space, whatever the investment may be.

[00:16:46] Speaker 00:
S&P 500, DK index, figures 15 through 17 have a number of different investments identified there.

[00:16:53] Speaker 00:
Pick the sample space.

[00:16:55] Speaker 00:
From within that sample space, let's use a bias parameter to affect the randomness of the resampling we're going to do.

[00:17:03] Speaker 00:
One way we can do that, set our bias parameter to 1.

[00:17:07] Speaker 00:
In that case, we're only going to draw updates from that investment, just the days that we're up.

[00:17:12] Speaker 00:
Or we could set our bias parameter to zero, in which case we're only going to select down days.

[00:17:19] Speaker 00:
Then we run our bootstrapping.

[00:17:20] Speaker 00:
We run our resampling on just the up days or just the down days.

[00:17:25] Speaker 00:
And in that way, we have biased our sample.

[00:17:29] Speaker 00:
We've affected the randomness in the resampling.

[00:17:32] Speaker 00:
We've biased our sampling.

[00:17:34] Speaker 00:
Sampling.

[00:17:35] Speaker 00:
And we have ended up with an analysis of bull and bear market conditions.

[00:17:41] Speaker 00:
That's what's discussed in the 291.

[00:17:42] Speaker 00:
Sortino.

[00:17:45] Speaker 00:
S&P 500 is an example of an asset that Sortino looked at, just one of nine, nine different assets.

[00:17:51] Speaker 00:
But he talks about 28 years of the S&P 500.

[00:17:53] Speaker 00:
We have that data.

[00:17:55] Speaker 00:
There's our sample space.

[00:17:57] Speaker 00:
From within that sample space, we can identify individual scenarios, seven different scenarios.

[00:18:03] Speaker 01:
But that's, let's see, this is, I guess, what I'm stuck on.

[00:18:05] Speaker 01:
It seems to me he's got seven sample spaces.

[00:18:08] Speaker 01:
That's what resampling would do.

[00:18:09] Speaker 01:
Each one

[00:18:10] Speaker 01:
He's created and we're going to do separate siloed resampling and we're going to get an expected value and a kurtosis or something.

[00:18:20] Speaker 01:
Whatever.

[00:18:23] Speaker 01:
And then we're going to take those numbers and we're going to create a linear combination of 20% of this and 30% of that and 2% of that and get a number that comes out of that.

[00:18:34] Speaker 00:
I think the confusion, Your Honor, is you're going to a next step.

[00:18:36] Speaker 00:
That's an additional

[00:18:37] Speaker 00:
disclosure, an additional embodiment that Sortino talks about, the waiting scenario.

[00:18:43] Speaker 01:
Do you think that the board relied on the waiting as part of its reasons?

[00:18:47] Speaker 00:
I do not.

[00:18:48] Speaker 00:
I think the board relied upon breaking down the sample space, the 28 years of the S&P 500.

[00:18:56] Speaker 01:
What was the language at A7 of both appendices in which the board said, the examiner said that the

[00:19:06] Speaker 01:
bias parameter doesn't have to be applied all at one time.

[00:19:10] Speaker 01:
It can include stuff after the sampling.

[00:19:12] Speaker 01:
Now the scenario definition is whatever else it is, it's not after the sampling, it's before the sampling.

[00:19:18] Speaker 01:
We would agree with that, right?

[00:19:20] Speaker 01:
So how did the board not rely on the weighting?

[00:19:23] Speaker 00:
But again, the sample space is selected.

[00:19:26] Speaker 00:
The S&P 500 sample space is selected first.

[00:19:29] Speaker 01:
Why are there not seven sample spaces?

[00:19:32] Speaker 00:
Those could be additional, they could be considered additional sample spaces, but what they also are are just like the up days and down days in the 291 patent.

[00:19:41] Speaker 00:
Just dividing that sample space up into bias parameter set to zero or bias parameter set to one, we've got individual samples then which we run the resampling on.

[00:19:50] Speaker 00:
That happens after in the 291.

[00:19:53] Speaker 01:
But the one and the zero and the minus one, as you've described it and I gather from the spec, has to do with what

[00:20:02] Speaker 01:
samples what you're drawing in the 4,000 iterations of sampling from each of the scenarios in its own silo.

[00:20:13] Speaker 01:
That's not what dividing the scenario, dividing the S&P results into seven scenarios is.

[00:20:21] Speaker 00:
It is, because we're dividing that S&P 500 sample into individual scenarios, individual monthly samples that correspond to that scenario.

[00:20:29] Speaker 00:
the months when we were in a deep recession, the months when we were in a real growth period, the months when we were in a mild recession period.

[00:20:37] Speaker 00:
It's breaking down that sample space into those individual scenarios and then running the resampling on those scenarios.

[00:20:42] Speaker 01:
What is the best evidence, declarations, I assume it would be, that says, I'm a statistician.

[00:20:50] Speaker 01:
I view the creation of these seven scenarios not as the definition of seven sample spaces, but as

[00:20:59] Speaker 01:
the process of sampling, because that distinction is actually in the claim between sample space and the selection of samples.

[00:21:10] Speaker 01:
What evidence says the selection of samples includes the definition of the scenarios?

[00:21:18] Speaker 00:
Just one clarification before I give you the record, Dr. Renner.

[00:21:20] Speaker 00:
Only claim one.

[00:21:21] Speaker 00:
Only claim one specifically calls out selecting a sample space, and then

[00:21:25] Speaker 00:
later on talks about the bias parameter in sample study.

[00:21:28] Speaker 00:
Not claim 11, not claim 29 doesn't even include the bias parameter.

[00:21:31] Speaker 00:
But I will cite you two of the declarations that the requester submitted to the office.

[00:21:37] Speaker 00:
And I'll cite you to Dr. Bloomfield's declaration.

[00:21:40] Speaker 00:
And that's at 657 through 58 of the record.

[00:21:44] Speaker 00:
And also Dr. Glasserman.

[00:21:46] Speaker 00:
That is at 703 to 04 of the record.

[00:21:50] Speaker 00:
And both of those

[00:21:52] Speaker 00:
declarations explain why a person of ordinary skill in the art would have understood Sortino to be describing the individual scenarios, bootstrapping those individual scenarios to correspond to what's disclosed in the 291.

[00:22:03] Speaker 01:
Do they say that a person of skill in the art would have understood the definition of the scenarios as biasing in the selection of samples with reference to that language?

[00:22:16] Speaker 03:
Council, don't shake your head up and down on sideways.

[00:22:20] Speaker 00:
They do not

[00:22:21] Speaker 00:
to address the in-sample selection, that specific language of the claim, Your Honor, because that was not added until after they had submitted their declarations.

[00:22:29] Speaker 01:
So we don't really have evidence about how a person of ordinary skill in the art would apply the language that is now at issue to this, what feels to me, quite crucial question of interpreting Sortino.

[00:22:44] Speaker 00:
But the addition of the in-sample selection

[00:22:47] Speaker 00:
It's a broad limitation.

[00:22:50] Speaker 00:
As Judge Clevenger was pointing out, that doesn't say only when you are selecting individual monthly samples or individual daily samples.

[00:22:59] Speaker 00:
It's not written that narrowly.

[00:23:01] Speaker 00:
It just says a bias parameter that determines the degree of randomness in sample selection in a resampling process.

[00:23:08] Speaker 00:
That's what happens in Sortino.

[00:23:12] Speaker 05:
But that depends on what meaning you read into the word in.

[00:23:17] Speaker 05:
And under the standard... I don't know what's the correct interpretation of the word in in that claim.

[00:23:24] Speaker 00:
We have to rely upon the broadest reasonable interpretation in the re-examination, and we submit that the broadest reasonable interpretation of in-sample selection in a re-sampling process was met by Sortino.

[00:23:35] Speaker 05:
Breaking down those individuals... It just means in general connection with at any point in time in the process.

[00:23:42] Speaker 00:
That sounds consistent to the... That's what you're hanging your hat on, right?

[00:23:45] Speaker 01:
Yes.

[00:23:46] Speaker 01:
Can you refresh my recollection of the prosecution history that led to the introduction of that language into the claims?

[00:23:53] Speaker 00:
The in-sample selection was added during the re-examination.

[00:23:56] Speaker 01:
During the re-examination.

[00:23:58] Speaker 01:
After a rejection that said what?

[00:24:00] Speaker 01:
Because that might be informative about what that term means.

[00:24:03] Speaker 00:
Rejection based upon Sortino as the primary reference.

[00:24:06] Speaker 00:
Sortino being the reference that

[00:24:08] Speaker 00:
broke up the individual scenarios, broke up the individual economic scenarios, and the patent owner made the argument to the office that we've added in samples.

[00:24:17] Speaker 05:
So they're making that amendment in order to get away from Sortino, right?

[00:24:22] Speaker 00:
That is the argument they made to the patent office, and the examiner correctly rejected it as did the board, because that broad language in sample selection did not distinguish Sortino.

[00:24:33] Speaker 00:
Did not distinguish Sortino.

[00:24:41] Speaker 05:
Why doesn't that point in the direction of saying that N means something very specific?

[00:24:46] Speaker 05:
Because if N has a general meaning, then Sortino would read on the claim.

[00:24:53] Speaker 05:
And they're trying to avoid having Sortino on the claim.

[00:24:56] Speaker 05:
That's why they went from one investment to two.

[00:25:00] Speaker 05:
Because they thought if we go to one investment instead of two, we'll get away from Sortino.

[00:25:06] Speaker 05:
Well, they only added the two.

[00:25:07] Speaker 05:
An argument is made in the companion case by the patent office that you can't take the file history of a re-examine into consideration in claim construction.

[00:25:16] Speaker 05:
For some reason, that's a no-no.

[00:25:20] Speaker 05:
Is that right?

[00:25:23] Speaker 00:
Well, I'm not recalling the specific argument that the office may have made in the other issue.

[00:25:26] Speaker 05:
I'm just following up on what I understood Judge Toronto's point to be.

[00:25:31] Speaker 00:
I would think that part of this court's consideration

[00:25:35] Speaker 00:
construction issues had they been raised.

[00:25:38] Speaker 00:
And again, our point is that these claim construction issues were not raised.

[00:25:42] Speaker 00:
But the court would be able to consider the entire record, including the record before the office.

[00:25:49] Speaker 00:
I think to the extent the Patent Number is trying to make a disclaimer argument based on what's happening live during the re-examination process, to the extent they're trying to read additional limitations into the claim saying, look, we amended the claim to add in-sample selection

[00:26:05] Speaker 00:
And we argued to the patent office that this is why it's distinguished over Sortino and other references.

[00:26:11] Speaker 05:
If they've written in their final history, well, we're looking at Sortino, and Sortino has a bias parameter, but the bias parameter is sort of not very specifically identified where it's happening.

[00:26:21] Speaker 05:
We're going to tell you exactly where ours is to distinguish over Sortino.

[00:26:26] Speaker 05:
And Sortino teaches one investment, and we're going to do two, because that'll make us different.

[00:26:30] Speaker 05:
That would be clear disclaimer, wouldn't it?

[00:26:35] Speaker 00:
Perhaps, Your Honor, although here they only added the two or more investments again to one of the claims.

[00:26:39] Speaker 05:
But the file history has to have some pertinent short of just claiming.

[00:26:42] Speaker 05:
We look at file history all the time to help us understand the meaning of claims.

[00:26:48] Speaker 00:
True.

[00:26:49] Speaker 00:
But I would think there's a difference between a patent owner or an applicant in original prosecution making an amendment, making an argument, having that argument be interpreted later on after the patent was set, after the patent had issued

[00:27:05] Speaker 00:
and trying to understand the meaning of an amendment that was made.

[00:27:10] Speaker 00:
Here, the process is still ongoing.

[00:27:12] Speaker 00:
They made this amendment in the reexamination, and they're saying to you, we added in sample selection to this claim, for example, on the bias parameter.

[00:27:21] Speaker 00:
And we also argued to the patent office all these other things.

[00:27:24] Speaker 00:
We now want you, and guess what?

[00:27:26] Speaker 00:
The board made a mistake because they didn't construe the claim to include all of our arguments.

[00:27:30] Speaker 00:
They want you to change the claim construction, adopt a new claim construction they never offered to the board,

[00:27:35] Speaker 00:
and read in all these limitations from the specification, limitations not supported by the claims, limitations that are inconsistent with the claims.

[00:27:42] Speaker 00:
They want you to read all that into the claims now.

[00:27:45] Speaker 00:
The cases we cited in our red brief, if they wanted those limitations in the claims, they should have amended them in the reexamination when they had the opportunity.

[00:27:53] Speaker 01:
Do you have anything you want to quickly say about the board's reasoning on the argument that whatever else Sortino shows, it does not show

[00:28:04] Speaker 01:
analyzing two or more investments in response to a single request.

[00:28:08] Speaker 00:
And our position, Your Honor, is that Sortino does disclose multiple investments, including the S&P 500 as one of the nine different asset categories that they studied.

[00:28:20] Speaker 00:
We think there was substantial evidence to support it.

[00:28:22] Speaker 01:
But where does Sortino disclose, even arguably, analyzing two or more investments in response to a single request?

[00:28:32] Speaker 01:
It's not even clear.

[00:28:35] Speaker 01:
In fact, I guess I will say it seems to me the board did not find that it did.

[00:28:39] Speaker 01:
The board relied on comprising to say one request, two requests, you can have as many requests as you want.

[00:28:44] Speaker 00:
I think that's an accurate summary of what the board did, Your Honor.

[00:28:47] Speaker 01:
That can't be what comprising does here.

[00:28:51] Speaker 00:
No, but the claim limitation reads a statistical analysis request corresponding to or pertaining to the two or more investors.

[00:28:59] Speaker 00:
So it's not just the comprising language.

[00:29:00] Speaker 00:
It's the open-ended A, right?

[00:29:03] Speaker 00:
A under the course case law is generally understood to be open-ended.

[00:29:06] Speaker 00:
It doesn't necessarily mean one and only one, as the patent owner here is arguing.

[00:29:10] Speaker 00:
And really what that argument becomes, Your Honor, the A statistical analysis request has to correspond to two investments.

[00:29:16] Speaker 00:
That necessarily requires the simultaneous analysis.

[00:29:18] Speaker 00:
That's really what they're getting at.

[00:29:19] Speaker 00:
In fact, my friend for the patent owner stood up this morning and he started explaining the invention as talking about temporal correlation and simultaneous analysis.

[00:29:28] Speaker 00:
None of those limitations are in the claim.

[00:29:30] Speaker 00:
None of those limitations are in the claim.

[00:29:32] Speaker 00:
the board correctly found, and there's substantial evidence to support the board, that serial requests satisfied the claim.

[00:29:40] Speaker 00:
Thank you, Your Honor.

[00:29:42] Speaker 00:
I'll just point out that one of the other points that the patent owner had made, that the board and the examiner had failed to appropriately consider the expert testimony, and I just want to point the court to our red brief, pages 48 through 49 and 58 through 62, where we walk through in detail

[00:30:01] Speaker 00:
where the board and the examiner actually did consider that evidence carefully, comprehensively, and completely.

[00:30:07] Speaker 00:
This is a case that the board got right, the examiner got right.

[00:30:10] Speaker 00:
We're asking the court to affirm in full.

[00:30:12] Speaker 00:
Thank you.

[00:30:19] Speaker 03:
Counselor, the reason I interrupted your opposing counsel was because you were shaking your head in denial when he was saying things, and that's not right.

[00:30:28] Speaker 02:
I'm sorry, Your Honor.

[00:30:32] Speaker 02:
Just responding to that.

[00:30:33] Speaker 02:
Go ahead.

[00:30:33] Speaker 02:
I'm sorry, Your Honor.

[00:30:36] Speaker 01:
Can you address, I'm sorry, put aside bias parameter, the two or more investments, put aside comprising.

[00:30:42] Speaker 01:
What about the indefinite article A?

[00:30:46] Speaker 02:
All by itself.

[00:30:47] Speaker 02:
Yes.

[00:30:48] Speaker 02:
A, Your Honor, based on prior jurisprudence from this court, could mean one or more than one, based on the context in which it's used.

[00:30:56] Speaker 02:
Here, you have claim, for example, claim 29.

[00:30:59] Speaker 02:
where you have A statistical analysis request is then following the same claim by D statistical analysis request within the same claim as an example showing that A means one and only one.

[00:31:15] Speaker 02:
This case is much more similar to a case like Harari versus Lee where you have a single bit line accessing a number of cells.

[00:31:24] Speaker 02:
Here you have a single statistical request corresponding to multiple investments.

[00:31:28] Speaker 02:
And so it's based on the context, Your Honor.

[00:31:32] Speaker 02:
A couple other really quick points on the waiver.

[00:31:35] Speaker 02:
The exact arguments were made in the course of prosecution, in the inter-parties re-examination, as is normally done in an inter-parties re-examination, which is that you make the argument, you don't have a separate section saying claim construction, but rather you argue what the terms mean

[00:31:52] Speaker 02:
in light of the prior art.

[00:31:54] Speaker 02:
And that was exactly done.

[00:31:55] Speaker 02:
The exact same arguments were made.

[00:31:57] Speaker 02:
And I previously pointed out where in the record it shows that.

[00:32:02] Speaker 02:
One additional point.

[00:32:04] Speaker 02:
The in-sample selection language was introduced.

[00:32:09] Speaker 02:
And you can see this on page 265 of the record.

[00:32:11] Speaker 02:
It was actually introduced in response to a comment made by appellees in prosecution.

[00:32:18] Speaker 02:
What they said was, if you really meant in-sample selection, then put it in the

[00:32:22] Speaker 02:
That's what they said on page 266, and we did it.

[00:32:25] Speaker 02:
We did it.

[00:32:26] Speaker 01:
That's exactly what we did.

[00:32:27] Speaker 01:
They asked for it.

[00:32:28] Speaker 01:
Is there something about what they said that helps us answer the question whether that meant a step after the scenario definition?

[00:32:46] Speaker 01:
Besides just the language by itself, something that says,

[00:32:49] Speaker 01:
Now this really means, after you've defined your sample space, and that includes the scenario definition, it means the picking of the thousand, I don't know what the numbers are, but the many, many, many samples on which you then do your

[00:33:02] Speaker 01:
And that's exactly the process.

[00:33:06] Speaker 01:
My question was about the record.

[00:33:09] Speaker 01:
Is there something that helps us see whether that distinction was being made or not?

[00:33:14] Speaker 01:
Yes.

[00:33:15] Speaker 05:
Tell us again what pages in the record.

[00:33:19] Speaker 02:
On the record from page 1245 to 1247 and pages 1372 to 1375,

[00:33:27] Speaker 02:
We have a discussion of the bias parameter and exactly how that works and how in-sample... I'm talking about the in-sample selection amendment.

[00:33:34] Speaker 05:
That's what Judge Toraya was focusing on.

[00:33:36] Speaker 05:
Where's the specific reference in the record that will teach us that you gave a specific meaning to the word in in response to, in making the amendment in response to an objection?

[00:33:51] Speaker 02:
I believe, Your Honor, these citations correspond to the introduction of the amendments made during prosecution in pages 1245 to 1247 and 1372 to 1375, where we made the precise arguments on those changes.

[00:34:10] Speaker 02:
Your Honor, just a couple of other very quick... We'll wrap it up because you're over your time.

[00:34:14] Speaker 02:
Okay, thank you.

[00:34:15] Speaker 02:
Your Honor, BRI cannot mean any possible construction.

[00:34:18] Speaker 02:
It still has to mean something reasonable.

[00:34:20] Speaker 02:
And all of this

[00:34:21] Speaker 02:
would not have occurred if the board had done a claim construction.

[00:34:24] Speaker 02:
We wouldn't be arguing about bias parameter, et cetera.

[00:34:28] Speaker 02:
And meaningful appellate review is not possible because the board has not bothered to construe the claims.

[00:34:33] Speaker 02:
And in fact, if you look at page six of the opinion, you'll see that their construction of bias parameter is contrary.

[00:34:41] Speaker 05:
Did you ask them to construe the claim?

[00:34:44] Speaker 05:
Your honor, we clearly told them that this... I know, but you're now complaining that the board didn't construe the claim.

[00:34:49] Speaker 05:
We're in a case here where your adversary says you didn't even make an argument on claim construction.

[00:34:53] Speaker 05:
How can you fault the board for not having done something in this scenario you're talking about?

[00:34:59] Speaker 02:
No, your honor, we absolutely told the board this is what the claim terms mean.

[00:35:03] Speaker 03:
Did you ask them to construe the claim?

[00:35:07] Speaker 05:
What we did was we argued... Yes, I'd say yes or no answer.

[00:35:11] Speaker 05:
In the record where you said, board, please consider the claim.

[00:35:15] Speaker 02:
Because I can't find it in what you're citing.

[00:35:17] Speaker 02:
What we did was we did not ask for a specific construction per se.

[00:35:23] Speaker 02:
What we did was we argued for the meaning of those terms.

[00:35:28] Speaker 02:
So that's our note.

[00:35:30] Speaker 02:
Your Honor, it wasn't per se.

[00:35:32] Speaker 02:
You're right.

[00:35:33] Speaker 03:
But what we did- Your time's up, counsel.

[00:35:34] Speaker ?:
Thank you.