`Tel: 571-272-7822
`
`
`Paper 52
`Entered: May 28, 2015
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`_______________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`_______________
`
`INTERNATIONAL BUSINESS MACHINES CORPORATION,
`Petitioner,
`
`v.
`
`INTELLECTUAL VENTURES II LLC,
`Patent Owner.
`_______________
`
`Case IPR2014-00180
`Patent 7,634,666 B2
`_______________
`
`
`
`Before MIRIAM L. QUINN, DAVID C. MCKONE,
`and JAMES A. TARTAL, Administrative Patent Judges.
`
`McKONE, Administrative Patent Judge.
`
`DECISION
`ON REQUEST FOR REHEARING
`37 C.F.R. § 42.71
`
`
`
`
`
`
`
`
`
`
`
`IPR2014-00180
`Patent 7,634,666 B2
`
`
`I. INTRODUCTION
`
`International Business Machines Corp. (“Petitioner”) filed a Petition
`
`(Paper 1, “Pet.”) to institute an inter partes review of claims 1–11 of U.S.
`
`Patent No. 7,634,666 B2 (Ex. 1005, “the ’666 patent”). Intellectual
`
`Ventures II LLC (“Patent Owner”) filed a Patent Owner Response (Paper
`
`24, “PO Resp.”) and Petitioner filed a Reply to the PO Response (Paper 29,
`
`“Reply”). We conducted an oral hearing on January 13, 2015 (Paper 49,
`
`“Tr.”). In a Final Written Decision (Paper 50, “Dec.”), we held that
`
`Petitioner failed to show, by a preponderance of the evidence, that any
`
`challenged claim was unpatentable as obvious over Matsuzaki (Ex. 1008)
`
`and Dworkin (Ex. 1012).
`
`In a Request for Rehearing (Paper 51, “Req.”), Petitioner contends
`
`that (1) we construed the term “feedback” too narrowly; (2) Dworkin teaches
`
`“feedback” under the broader construction; and (3) we misapprehended or
`
`overlooked Petitioner’s evidence of reasons to combine Matsuzaki and
`
`Dworkin. As detailed below, we are not persuaded by Petitioner’s
`
`arguments.
`
`
`
`II. ANALYSIS
`
`The burden of showing that the Decision should be modified is on
`
`Petitioner, the party challenging the Decision. See 37 C.F.R. § 42.71(d). In
`
`addition, “[t]he request must specifically identify all matters the party
`
`believes the Board misapprehended or overlooked, and the place where each
`
`matter was previously addressed in a motion, an opposition, or a reply.” Id.
`
` 2
`
`
`
`
`
`
`
`
`
`IPR2014-00180
`Patent 7,634,666 B2
`
`
`A. Construction of “feedback”
`
`Claim 1 recites “wherein the outputs of the multiplication unit, the
`
`addition unit and the sign inversion unit are feedback to the arithmetic
`
`controller.” Independent claim 4 includes a similar recitation.
`
`In the Petition, Petitioner proposed construing “feedback” to mean “a
`
`result that is directly transmitted back.” Pet. 8. Petitioner essentially argued
`
`that “feedback” should be limited to an example described in the ’666 patent
`
`in which each of a multiplication unit, an addition unit, and a sign inversion
`
`unit sends its own interim result directly back to a controller without passing
`
`through intermediate components. Pet. 8–9 (citing Ex. 1005, 3:30–39).
`
`Patent Owner opposed Petitioner’s construction in the Preliminary
`
`Response (Paper 9, “Prelim. Resp.”) at 3, arguing that Petitioner was reading
`
`a limitation into the claims from the Specification. In the Decision to
`
`Institute (Paper 10, “Inst. Dec.”) at 11, we declined to limit “feedback” to a
`
`disclosed embodiment in which results are fed back directly. In its PO
`
`Response, Patent Owner agreed with our analysis. PO Resp. 9. Petitioner
`
`did not argue the construction of this term further in the Reply.
`
`At the oral hearing, the panel challenged Patent Owner with questions
`
`on the issue of whether the claims contemplate either direct or indirect
`
`feedback. Tr. 44:1246:2. Patent Owner conceded that the claims do not
`
`require either direct or indirect feedback, and that in the instance of indirect
`
`feedback, intermediate components should not change the content of the data
`
`being sent back to the controller, for example latches and multiplexers that
`
`would delay, but not change, the data sent back to the controller. Tr. 44:12–
`
`46:2. Patent Owner conceded that if an intermediate component changed
`
` 3
`
`
`
`
`
`
`
`IPR2014-00180
`Patent 7,634,666 B2
`
`
`the data output by an arithmetic device, the feedback would no longer be the
`
`output of that device. Id. (Patent Owner arguing that Figure 2 of the ’666
`
`patent supports its contention that the output that constitutes feedback under
`
`the claims does not enter another addition or sign inversion unit before it is
`
`provided back). Based on the evidence and arguments presented in the
`
`Petition and the PO Response, we construed “wherein the outputs of the
`
`multiplication unit, the addition unit and the sign inversion unit are feedback
`
`to the arithmetic controller” to mean that the output values of the
`
`multiplication unit, the addition unit, and the sign inversion unit are
`
`feedback to the arithmetic controller, although those values may pass,
`
`unchanged, through intermediate components (e.g., latches and
`
`multiplexers). Dec. 9.
`
`In the Request, Petitioner contends that Patent Owner “changed
`
`course” and introduced a new claim construction argument at the hearing.
`
`Req. 2. We disagree. Patent Owner responded to our questions seeking to
`
`focus the dispute between the parties as reflected in the Petition and PO
`
`Response. The construction of “feedback” in our Decision was not the
`
`acceptance of an improperly raised argument. Rather, our construction was
`
`based upon the evidence presented in the Petition and the PO Response.
`
`Petitioner argues that it objected to Patent Owner’s purportedly new
`
`construction of “feedback.” Req. 2–3. We do not agree with Petitioner. At
`
`the hearing, Petitioner was instructed to raise its objections during the time it
`
`was allotted for its argument. Tr. 4:14–17. Petitioner argued that it
`
`disagreed with how Patent Owner interpreted the alleged feedback operation
`
`of Dworkin, and referred to its detailed “step-by-step graphics” presented on
`
` 4
`
`
`
`
`
`
`
`IPR2014-00180
`Patent 7,634,666 B2
`
`
`rebuttal in Petitioner’s Reply. Tr. 10:2311:9. After its rebuttal time had
`
`expired, Petitioner requested (and was granted) additional time to raise an
`
`objection “to all of the detailed Dworkin argument that was offered” by
`
`Patent Owner, arguing that it was “not anywhere present in [Patent Owner’s]
`
`opposition.” Tr. 61:13–62:7. We are unable to discern how Petitioner’s
`
`generic objection to Patent Owner’s detailed explanation of Dworkin was
`
`directed to the purportedly new claim construction argument made by Patent
`
`Owner at the hearing. Petitioner has failed to point to any objection to
`
`Patent Owner’s answer to our question or any discussion of the construction
`
`of “feedback” in its main or rebuttal argument. In any event, to the extent
`
`Petitioner believes it properly objected to Patent Owner’s candid response to
`
`the panel’s questions regarding the scope of “feedback,” that objection is
`
`overruled.
`
`In sum, we are not persuaded that we overlooked or misapprehended
`
`the evidence and arguments presented regarding the construction of
`
`“feedback,” and find no reason to modify our construction of “feedback” in
`
`the Decision.
`
`
`
`B. We did not misapprehend or overlook Dworkin’s teaching of
`feedback
`
`Petitioner argues that our construction of “feedback” was unfair
`
`because Petitioner demonstrated in the Petition and the Reply that Dworkin
`
`teaches indirect “feedback” consistent with the reasoning in our Decision to
`
`Institute. Req. 2. Petitioner argues that our preliminary construction “[does
`
`not] require that output values must be feedback to the controller
`
`‘unchanged.’” Id. at 5. According to Petitioner, it presented evidence that
`
` 5
`
`
`
`
`
`
`
`IPR2014-00180
`Patent 7,634,666 B2
`
`
`output values from Dworkin’s sub-ALUs are fed to intermediate
`
`components, that those intermediate components change the values of the
`
`sub-ALU outputs, and that those changed values eventually reach the
`
`controller. Id. at 6–7. We disagree that Petitioner presented any such
`
`evidence or argument in either the Petition or the Reply. Rather, this is a
`
`third, new theory of Dworkin’s operation presented for the first time in the
`
`Request.
`
`Petitioner first contended, in the Petition, that Dworkin teaches
`
`directly feeding back the outputs of a multiplication unit, a sign inversion
`
`unit, and an addition unit to a controller. Pet. 30. Petitioner did not argue
`
`that feedback could include data that changed in value from an arithmetic
`
`unit to a controller. Such an argument would have been inconsistent with
`
`Petitioner’s proposed construction that feedback must be direct. Nor did
`
`Petitioner introduce evidence that Dworkin’s sub-ALUs feed their results to
`
`adjacent sub-ALUs, which change the results and pass them, eventually, to a
`
`controller. See Ex. 1001 ¶¶ 116–18.
`
`In the PO Response, Patent Owner introduced evidence that Dworkin
`
`does not teach a sign-inversion unit, for example; thus, it cannot teach
`
`feeding back (directly or indirectly) the output of a sign-inversion unit to a
`
`controller. Dec. 16; PO Resp. 13. Patent Owner also introduced evidence
`
`(summarized at Dec. 17–19) that the output of each of Dworkin’s sub-ALUs
`
`(save the left-most sub-ALU) is fed to an adjacent sub-ALU and is not fed
`
`back to the controller.
`
`In the Reply, Petitioner changed its theory on how Dworkin operates.
`
`First, Petitioner admitted that Dworkin in fact did not teach feeding back the
`
`outputs of a multiplication unit, a sign inversion unit, and an addition unit to
`
` 6
`
`
`
`
`
`
`
`IPR2014-00180
`Patent 7,634,666 B2
`
`
`a controller. Reply 7. Petitioner then introduced a second theory of
`
`Dworkin’s operation, supported by a new declaration from its expert. In
`
`particular, the Reply contended that Patent Owner’s description of Dworkin
`
`was incorrect and that the outputs of each of the sub-ALUs themselves (not
`
`values changed by intermediate calculations) were left-shifted to the
`
`controller. Reply 8–9.
`
`Upon consideration of all the evidence, including that presented in the
`
`Petition and the Reply, we found that the evidence presented by Patent
`
`Owner was much more consistent with the actual disclosure of Dworkin and,
`
`thus, credited Patent Owner’s evidence. Specifically, we weighed the
`
`evidence and made a factual finding that Patent Owner’s expert testimony
`
`was credible, accurately describing Dworkin, and that Petitioner’s expert
`
`testimony lacked credibility, inaccurately describing Dworkin. Dec. 20–21.
`
`The argument Petitioner now presents in the Request was not
`
`presented in either the Petition or the Reply and is not supported by either
`
`declaration of its expert (the first declaration was directed to the first theory
`
`in the Petition and the Reply declaration was directed to the second theory in
`
`the Reply). A request for rehearing is not an opportunity to expand on
`
`evidence and arguments not presented, or to mend gaps in the evidence
`
`relied on in the Petition. We could not have misapprehended or overlooked
`
`Petitioner’s third theory regarding Dworkin’s operation when argument and
`
`evidence supporting such a theory was not developed sufficiently in the
`
`Petition or the Reply. Moreover, this new theory is inconsistent with
`
`Petitioner’s contention in the Petition that Dworkin “describes several
`
`computation arithmetic units, where the output of each unit is directly sent
`
`back to the arithmetic controller.” Pet. 17. In sum, we are not persuaded
`
` 7
`
`
`
`
`
`
`
`IPR2014-00180
`Patent 7,634,666 B2
`
`
`that we overlooked or misapprehended the evidence and arguments
`
`presented regarding Dworkin’s teaching of “feedback,” and find no reason to
`
`modify our Decision.
`
`
`
`C. We did not misapprehend or overlook Petitioner’s evidence of a
`reason to combine Matsuzaki and Dworkin
`
`In our Decision, we considered the evidence presented in the Petition,
`
`the Reply, and the corresponding expert declarations, and found that
`
`Petitioner’s stated reasons to combine Matsuzaki and Dworkin amounted to
`
`nothing more than that those references had substantial similarities.
`
`Dec. 21–29. We concluded that these reasons to combine were insufficient
`
`to show obviousness. Id. at 29–30. As to Petitioner’s “common sense”
`
`argument, after weighing the evidence, we found a lack of evidentiary
`
`support. Id. at 26.
`
`Petitioner first argues that we improperly required Petitioner to
`
`explain how Matsuzaki and Dworkin would have been combined. Req. 9.
`
`Petitioner is correct that the law does not require a showing that Matsuzaki
`
`and Dworkin could be bodily incorporated into one another. See In re
`
`Mouttet, 686 F.3d 1322, 1332 (Fed. Cir. 2012). We imposed no such
`
`requirement. Rather, in evaluating the scope and content of the prior art, we
`
`observed that the evidence presented in the Petition failed to show how or
`
`why the teachings of Dworkin were being applied to those of Matsuzaki.
`
`Dec. 16–17. Specifically, as explained above, the Petition contended that
`
`Dworkin disclosed directly feeding back the outputs of multiplication, sign-
`
`inversion, and addition units to a controller. Petitioner later admitted that
`
`this explanation of Dworkin was incorrect and advanced a new theory of
`
` 8
`
`
`
`
`
`
`
`IPR2014-00180
`Patent 7,634,666 B2
`
`
`Dworkin’s operation. Reply 7. Our Decision found that the Petition,
`
`because it was based on the first theory, lacked an explanation of how or
`
`why a skilled artisan would have modified Matsuzaki under Petitioner’s
`
`second theory of Dworkin’s teachings. Dec. 16–17.
`
`Petitioner next lists several arguments that it alleges it presented
`
`during the trial and that we overlooked or misapprehended. For example,
`
`Petitioner contends that we overlooked evidence in the Petition that Dworkin
`
`teaches feeding back the output of generic computational units to a
`
`controller. Req. 10. The Petition is unambiguous as to Petitioner’s
`
`contention that Dworkin teaches feedback: “Dworkin discloses wherein the
`
`outputs of the multiplication unit (e.g., sub-ALU 18), the addition unit (e.g.,
`
`sub-ALU 18) and the sign inversion unit (e.g., sub-ALU 18) are directly sent
`
`back to the arithmetic controller (e.g., controller 20).” Pet. 30. Petitioner
`
`did not argue in the Petition that Dworkin teaches feeding back the outputs
`
`of generic computational units. Petitioner only introduced that argument in
`
`the Reply, after conceding the argument presented in the Petition.
`
`Petitioner contends that we overlooked evidence in the Petition that
`
`pipeline feedback (purportedly described in Matsuzaki) and cascade
`
`feedback (purportedly described in Dworkin) were “known and predictable
`
`solutions, i.e., alternatives.” Req. 10–11 (citing Pet. 15; Ex. 1001 ¶¶ 42–43,
`
`118). Petitioner’s evidence, at most, purports to show that the Examiner,
`
`during prosecution of the ’666 patent, considered pipeline feedback and
`
`cascade feedback to be different and patentably distinct. Ex. 1001 ¶¶ 42–43.
`
`Petitioner did not argue that pipeline and cascade feedback were known
`
`alternatives or, for that matter, known to be interchangeable. Thus, we could
`
`not have overlooked or misapprehended this argument. Similarly, Petitioner
`
` 9
`
`
`
`
`
`
`
`IPR2014-00180
`Patent 7,634,666 B2
`
`
`did not argue that Dworkin’s feedback was an improvement that could have
`
`been applied predictably to Matsuzaki (Req. 11–12). We could not have
`
`overlooked or misapprehended that argument either.
`
`Petitioner also contends that we overlooked evidence that a skilled
`
`artisan would have combined Matsuzaki and Dworkin to “control how
`
`computations are performed by the arithmetic unit.” Req. 12 (citing
`
`Ex. 1001 ¶ 148). Similarly, Petitioner argues that we overlooked evidence
`
`that Matsuzaki and Dworkin both address optimal hardware
`
`implementations. Id. at 12–13 (citing Ex. 1001 ¶ 145). Petitioner also
`
`contends that we “summarily dismissed” evidence that the presence of
`
`Montgomery reduction in both references would have led a skilled artisan to
`
`consider Matsuzaki and Dworkin together. Id. at 13 (citing Ex. 1001 ¶ 146).
`
`Petitioner’s arguments are not persuasive. We considered each of these
`
`arguments and the corresponding declaration testimony in detail in our
`
`Decision, at 24–25. We found that this evidence showed nothing more than
`
`that the references had similarities. Id. Mere disagreement with our
`
`Decision is not grounds for rehearing.
`
`Petitioner further argues that a prior panel of this Board found
`
`“similarities between the art alone” to be sufficient to show a reason to
`
`combine, a decision we should follow in this proceeding to avoid being
`
`arbitrary and panel-specific. Req. 14–15 (citing Toyota Motor Co. v. Am.
`
`Vehicular Sci. LLC, case IPR2013-00414, 25–26 (Jan. 6, 2015) (Paper 45)).
`
`The decision to which Petitioner cites, however, also considered evidence
`
`that a non-interactive display was known to be interchangeable with an
`
`interactive display and that the substitution would have been a simple design
`
`choice, a portion of the decision Petitioner improperly crops out with
`
`
`
`
`10
`
`
`
`IPR2014-00180
`Patent 7,634,666 B2
`
`
`ellipses in its explanation. When considered in its entirety, Toyota is
`
`consistent with our discussion of Tyco Healthcare Group LP v. Ethicon
`
`Endo-Surgery, Inc., 774 F.3d 968, 977–79 (Fed. Cir. 2014). Dec. 28–29.
`
`As explained in the Decision, Petitioner did not argue, nor present evidence
`
`to support an argument, that the feedback techniques of Matsuzaki and
`
`Dworkin were known to be interchangeable. Id. Thus, we could not have
`
`overlooked or misapprehended such an argument.
`
`Petitioner has not shown that we overlooked or misapprehended the
`
`evidence and arguments presented regarding a reason to combine Matsuzaki
`
`and Dworkin, and, therefore, we find no reason to modify our Decision.
`
`
`
`III. CONCLUSION
`
`Petitioner has not shown that we misapprehended or overlooked any
`
`of its arguments or evidence, or that we misapplied the law. Accordingly,
`
`we decline to change our Decision.
`
`IV. ORDER
`
`Petitioner’s Request for Rehearing is denied.
`
`
`
`PETITIONER:
`
`Kenneth R. Adamo
`Eugene Goryunov
`KIRKLAND & ELLIS LLP
`kenneth.adamo@kirkland.com
`eugene.goryunov@kirkland.com
`
`
`
`11
`
`
`
`
`
`IPR2014-00180
`Patent 7,634,666 B2
`
`
`PATENT OWNER:
`
`Herbert D. Hart III
`Jonathan R. Sick
`Peter J. McAndrews
`MCANDREWS, HELD & MALLOY, LTD.
`hhart@mcandrews-ip.com
`jsick@mcandrews-ip.com
`pmcandrews@mcandrews-ip.com
`
`
`
`12