United States Court of Appeals for the Federal Circuit

Transcription

1 , United States Court of Appeals for the Federal Circuit QUEENS UNIVERSITY AT KINGSTON, v. Appellant, SAMSUNG ELECTRONICS CO., LTD., SAMSUNG ELECTRONICS AMERICA, INC., Appellees. Appeals from the United States Patent and Trademark Office, Patent Trial and Appeal Board, Nos. IPR and IPR BRIEF FOR APPELLANT Ian B. Crosby Susman Godfrey LLP 1201 Third Avenue, Suite 3800 Seattle, Washington Telephone: (206) Robert Greene Sterne Michelle K. Holoubek Sterne, Kessler, Goldstein & Fox PLLC 1100 New York Avenue, NW Washington DC Telephone: (202) Shawn Daniel Blackburn Susman Godfrey LLP 1000 Louisiana Street Suite 5100 Houston, Texas Telephone: (713) Counsel for Appellant JANUARY 9, 2017

2 FORM 9. Certificate of Interest Form 9 Rev. 03/16 UNITED STATES COURT OF APPEALS FOR THE FEDERAL CIRCUIT Queen's University at Kingston v. Samsung Electronics Co., Ltd. Case No , CERTIFICATE OF INTEREST Counsel for the: (petitioner) (appellant) (respondent) (appellee) (amicus) (name of party) Queen's University at Kingston certifies the following (use "None" if applicable; use extra sheets if necessary): 1. Full Name of Party Represented by me Queen's University at Kingston 2. Name of Real Party in interest (Please only include any real party in interest NOT identified in Question 3) represented by me is: Queen's University at Kingston 3. Parent corporations and publicly held companies that own 10 % or more of stock in the party None Parteq Innovations Northwater Capital, Inc. 4. The names of all law firms and the partners or associates that appeared for the party or amicus now represented by me in the trial court or agency or are expected to appear in this court (and who have not or will not enter an appearance in this case) are: Sterne, Kessler, Goldstein & Fox P.L.L.C.: Lestin Kenton Jan 9, 2017 /s/ Ian B. Crosby Date Please Note: All questions must be answered cc: Counsel of Record Ian B. Crosby Signature of counsel Printed name of counsel Reset Fields

3 TABLE OF CONTENTS Page CERTIFICATE OF INTEREST... ii TABLE OF CONTENTS... ii TABLE OF AUTHORITIES... iv STATEMENT OF RELATED CASES... 1 JURISDICTIONAL STATEMENT... 2 STATEMENT OF THE ISSUES... 3 STATEMENT OF THE CASE... 4 I. THE 665 AND 660 PATENTS... 4 II. GOLDSTEIN III. INTER PARTES REVIEW A. Samsung s Initial Grounds of Unpatentability B. Queen s University s Patent Owner Responses C. Samsung s New Grounds of Unpatentability and New Evidence in Support of Its Replies D. The Board s Reliance on Samsung s New Evidence and New Ground of Unpatentability, and Faulting Queen s University for not Presenting Evidence the Board Explicitly Forbade ii

4 SUMMARY OF ARGUMENT ARGUMENT I. STANDARDS OF REVIEW II. III. THE BOARD ERRED BY RELYING ON NEWLY PRESENTED EVIDENCE AND NEW GROUNDS OF UNPATENTABILITY THE BOARD ERRED IN FINDING ANTICIPATION BY GOLDSTEIN A. Goldstein Does Not Disclose an Eye Sensor In or On a Device B. Goldstein Does Not Enable the Claims Hardware Sensor C. The Board s Decisions Cross the Line Between Anticipation and Obviousness IV. CONCLUSION AND STATEMENT OF RELIEF SOUGHT ADDENDUM CERTIFICATE OF SERVICE CERTIFICATE OF COMPLIANCE iii

5 TABLE OF AUTHORITIES CASES Page(s) Abbott Labs. v. Cordis Corp., 710 F.3d 1318 (Fed.Cir.2013) Amgen Inc. v. Hoechst Marion Roussel, Inc., 314 F. 3d 1313 (Fed. Cir. 2003) Belden Inc. v. Berk-Tek LLC, 805 F.3d 1064 (Fed. Cir. 2015)... 29, 30, 40 Carnegie Mellon Univ. v. Marvell Tech. Grp., Ltd., 807 F.3d 1283 (Fed. Cir.) Ciba-Geigy Corp. v. Alza Corp., 68 F.3d 487 (Fed. Cir. 1995) Connell v. Sears, Roebuck & Co., 722 F.2d 1542 (Fed.Cir.1983) Crown Operations International, Ltd. v. Solutia Inc., 289 F.3d 1367, 62 USPQ2d 1917 (Fed. Cir. 2002) Dell Inc. v. Acceleron, LLC, 818 F.3d 1293 (Fed. Cir. 2016)... passim Finisar Corp. v. DirecTV Group, Inc., 523 F.3d 1323 (Fed. Cir. 2008) Hockerson-Halberstadt, Inc. v. Avia Group Int l., 222 F.3d 951 (Fed. Cir. 2000) HP Inc. v. MPHJ Tech. Inv., LLC, 817 F.3d 1339 (Fed. Cir. 2016) iv

6 In re Antor Media Corp., 689 F.3d 1282 (Fed. Cir. 2012)... 10, 28 In re Arkley, 455 F.2d 586 (1972) In re Donohue, 766 F.2d... 36, 52 In re Magnum Oil Tools Int l, Ltd., 829 F.3d 1364 (Fed. Cir. 2016)... 29, 39 In re NuVasive, Inc., 841 F.3d 966 (Fed. Cir. 2016)... passim In re Omeprazole Patent Litig., 483 F.3d 1364 (Fed. Cir. 2007) In re Queen s University at Kingston, 820 F.3d 1287 (Fed. Cir. 2016)... 1 In re Wright, 569 F.2d 1124 (CCPA 1977) Microsoft Corp. v. Proxyconn, Inc., 789 F.3d 1292 (Fed. Cir. 2015) Net MoneyIN, Inc. v. VeriSign, Inc., 545 F.3d 1359 (Fed. Cir. 2008) On-Line Careline, Inc. v. Am. Online, Inc., 229 F.3d 1080 (Fed. Cir. 2000) Studiengesellschaft Kohle, m.b.h. v. Dart Indus., Inc., 726 F.2d 724 (Fed. Cir. 1984)... 52, 53, 54 Teleflex, Inc. v. Ficosa N. Am. Corp., 299 F.3d 1313 (Fed. Cir. 2002)... 36, 52, 54 v

7 STATUTES 5 U.S.C U.S.C. 554(b)(3) U.S.C. 554(c) U.S.C. 556(d)... 31, 38, 40 5 U.S.C U.S.C. 1295(a)(4)(A) U.S.C. 102(e)... 16, U.S.C. 103(a)... 16, U.S.C. 316(e) U.S.C vi

8 STATEMENT OF RELATED CASES The captioned cases are two consolidated appeals from two inter partes reviews ( IPRs ), IPR and IPR , of U.S. Patent Nos. 7,762,665 ( the 665 Patent ) and 8,096,660 ( the 660 Patent ) (collectively, the Patents ), respectively. No prior appeal from these IPRs was previously before this Court or any other court. The Patents have been asserted by Queen s University at Kingston and PARTEQ Research and Development Innovations ( Queen s University ) in Queen s University at Kingston, et al. v. Samsung Electronics Co., Ltd., et al., No. 2:14-cv JRG-RSP (E.D. Tex.). That case is currently stayed pending the resolution of the present IPRs. Queen s University previously filed for a writ of mandamus directing the United States District Court for the Eastern District of Texas to withdraw an order compelling the production of communications between Queen s University employees and their patent agents. This Court granted that petition in In re Queen s University at Kingston, 820 F.3d 1287 (Fed. Cir. 2016). 1

9 JURISDICTIONAL STATEMENT The Patent Trial and Appeal Board ( the Board ) issued its final written decisions on July 27, Queen s University timely filed its notices of appeal on September 28, This Court has jurisdiction under 28 U.S.C. 1295(a)(4)(A) and 35 U.S.C

10 STATEMENT OF THE ISSUES 1. Whether the Board violated Queen s University s due process rights under the Administrative Procedure Act ( APA ) by refusing to allow Queen s University to present rebuttal evidence to evidence relied upon by the Board and presented for the first time in Samsung s reply briefs. 2. Whether the Board erred in finding Claim 11 of the 665 Patent and Claim 14 of the 660 Patent anticipated by U.S. Patent App. No. 2003/ ( Goldstein ) where the Board relied on multiple references to supplement the teachings of Goldstein. 3

11 STATEMENT OF THE CASE The inventions of the Patents were created at a time when user interaction with technological devices, particularly mobile devices, was disruptive and inefficient. The Patents disclose specific methods and apparatus for improving the effectiveness of interaction with computers and other devices. These inventions are the product of research by Dr. Roel Vertegaal on Attentive User Interfaces ( AUI ) and device-initiated communications between humans and technology. At the most basic level, the Patents employ sensors that detect certain physical signs that a person is paying attention to a device, convert information derived from those sensors into a signal, then use that signal to alter the device s operations on the basis of the person s attention. The inventive devices respond to human attention by acting similarly to the way humans would perceive such cues, a significant advance in the art. I. THE 665 AND 660 PATENTS The claims and common specification of the 665 and 660 Patents disclose and claim methods and apparatus for improving human interaction with technological devices, and the claims at issue in this appeal focus specifically on using hardware sensors in or on cellular telephones, 4

12 allowing for such devices to accurately detect the attentive state of a user and perform an action in response. Prior to the invention, most attempts to improve man-machine interaction focused on how a user could control the machine. Appx160 at 1: For example, some attempts explored controlling a device using eye tracking using a location of the user s point of gaze as an input for performing various actions. Id. For example, on stationary systems where the relationship between a user and a computer was known and constant, techniques had been explored using eye tracking devices to track a user s eye gaze direction. Id. By identifying where on a screen a user was looking, such systems could perform computer functions such as controlling a cursor, controlling information transfer, downloading, scrolling on a computer, selecting an image, and enhancing the perceived video quality of a portion of a computer display corresponding to a user s gaze. Id. The inventors of the Patents sought to improve a user s experience with devices even further. Appx160 at 2:4-7. Specifically, the Patents provide a solution to the problem of intrusive devices by augmenting devices 5

13 with attentive user interfaces: user interfaces that negotiate the attention they receive from or provide to users by negotiations through peripheral channels of interaction. Appx162 at 5:2-6. Human groups react to nonverbal visual clues to facilitate an efficient process of conversational turn-taking. Appx162 at 5:6-13. As explained in the specification, an attentive user interface according to the invention applies... social rules to device-initiated interactions or communications, by assessing a user s attentive state, and making a determination as to whether, when, and how to interrupt (e.g., notify) the user on the basis of the user s attentive state. Appx162 at 5: The inventors applied these attentive user interfaces to devices ranging from furniture and appliances to computers and cellular telephones. Appx165 at 11:66-12:5. Their inventions are embodied in the claims, which each require a sensor in or on the device that senses attention toward the device, and the device specifically at issue in this appeal is a cellular telephone. For example, Claims 1 and 14 of the 660 Patent are drawn to: 1. Apparatus for communication between a user and a device, 1 comprising: 1 All emphasis added unless otherwise noted. 6

14 a hardware sensor in or on the device that senses attention of the user specifically toward the device; and a processor that processes a signal from the hardware sensor and outputs to the device a measure or index of the user s attention toward the device; and wherein the operation of the device is modulated on the basis of the measure or index of the user s attention toward the device; wherein the operation that is modulated is initiated by the device and provides a notification and/or information and/or communication to the user based on the user s attention toward the device. 14. The apparatus of claim 1, wherein the device is a cellular telephone. Appx170 at 22:22-35, Appx170 at 23:5-6. Similarly, Claims 1 and 11 of the 665 Patent recite: 1. A method of modulating operation of a device, comprising: disposing a hardware sensor in or on the device for sensing attention of a user specifically toward the device; processing a signal from the hardware sensor and outputting to the device a measure or index of the user s attention toward the device; and modulating operation of the device on the basis of the measure or index of the user s attention toward the device; wherein the operation that is modulated is initiated by the device and provides a notification and/or information and/or 7

15 communication to the user based on the user s attention toward the device. 11. The method of claim 1, wherein the hardware device is a cellular telephone. Appx150 at 21:50-63, Appx150 at 22: The specification distinguishes between different types of sensors and their associated measures of attention. For example, the specification discusses eye tracking sensors (or eye trackers) that detect eye movement and eye gaze direction to determine the direction a person is looking. Appx160 at 1:32-52, Appx163 at 7:3-4. The specification also discloses eye contact sensors, which are distinct from eye trackers. Appx at 6:67-7:4. Eye contact sensors detect eye contact when a subject or user is looking at the sensor, whereas eye trackers detect eye movement to determine the direction a subject or user is looking. Id. That is, an eye contact sensor detects when a user s eye is looking generally at the sensor, while an eye tracking sensor calculates the exact position in space where the eye is looking (i.e., the eye s point of gaze) in an area around the sensor. Importantly, eye gaze direction is detected using an eye tracking sensor rather than an eye contact sensor, because 8

16 an eye contact sensor on its own does not detect the necessary information to calculate the point and direction of gaze. Appx164 at 9: Eye trackers, in contrast, detect eye gaze angle, which is a two dimensional direction intersecting a two dimensional location on a screen, to determine the x,y location at which a subject or user is looking. Appx459, Appx1343, Appx , Appx Eye gaze direction measurement using a traditional eye tracking sensor necessitates calibration to match directional measurements to the planar intersection with the screen, whereas eye contact sensing does not require any calibration. Appx166 at 13: While the appealed claims require only a sensor, and do not explicitly recite either an eye contact sensor or an eye tracking sensor, the difference between the two is important. The alleged prior art, Goldstein (discussed below), only discloses an eye tracker. So, to be enabled, Goldstein must show how an eye tracker is used to meet the claimed combination. Specifically, Goldstein must enable an eye tracker in or on a cellular telephone. Further, the only device in Goldstein (and the one alleged by Samsung to be a cellular telephone is a mobile communication device. Appx , Appx , Appx , Appx

17 So, Goldstein must show how an eye tracker is implemented in or on a mobile cellular telephone. In re Antor Media Corp., 689 F.3d 1282, 1291 (Fed. Cir. 2012) (the prior art reference needs to enable the portions of its disclosure alleged to anticipate the claimed invention. ); Amgen Inc. v. Hoechst Marion Roussel, Inc., 314 F. 3d 1313, 1355 (Fed. Cir. 2003) ( A claimed invention cannot be anticipated by a prior art reference if the allegedly anticipatory disclosures cited as prior art are not enabled ). Yet eye tracking was no simple task in 2003, when the patent was filed. Appx , Appx , Appx , Appx The large amount of data needed to be processed by eye tracking systems meant that eye tracking typically required the use of desktop computing power. Id. Though some experts experimented with portable tracking systems, such systems were head-mounted eye-tracking systems that needed large backpacks to carry processors and batteries. Appx , Appx Head-mounted systems were chosen for such portable systems because of the difficulty of tracking the eye accurately through relative movement of the user and tracker an eye tracking sen- 10

18 sor could only produce accurate results if the geometry (distance and angle) between the user s pupil and the eye tracking sensor at the time of detection was known. Id, Appx , Appx As such, implementing an eye tracking sensor on a mobile device such as a cellular telephone presented many challenges, including mobility issues (e.g., changes in relative movement of user and camera), form factor of the device, and device resource limitations. Appx , Appx , Appx , Appx As late as 2007, Drewes et al. reported that eye-tracking technology for interaction with mobile phones is not yet available. Appx , Appx , Appx6205. Experiments with eye tracking sensors on mobile devices in a truly mobile environment were unsuccessful, as the distance between the mobile device and the user changed, resulting in a loss in calibration. Appx , Appx So, for such experiments, mobile devices were actually immobilized by locking them into one position, and the users heads were also immobilized by using a chin rest to stabilize head movement. Id. Such experiments also used eye trackers that were separated from the mobile devices, because the size of such eye trackers 11

19 prevented their incorporation into the small form factors of mobile devices such as cellular telephones. Id. This is in comparison to the Patents, which disclose miniaturized eye contact sensors that can be embedded in electronic devices, such as cellular phones, to measure a user s attentive state and control delivery of information accordingly. Appx at 14:34-15:67. No calibration was needed with an eye contact sensor, so the relative movement between a user and device did not present the same challenges. Appx166 at 13: The device may deliver information only when the user is paying attention to the device and defer that information s delivery when the user s attention is directed elsewhere. Appx167 at 15:4-37. This may include providing notifications to a user or the live pausing of audiovisual material. Appx167 at 15: For example, for content being displayed on a device which has live pausing capability and is augmented with an attentive user interface according to the claims, the device may pause programming when it senses that the user is not watching the device and resume the program when the user looks at the device. Appx168 at 17:

20 II. GOLDSTEIN Goldstein describes an apparatus for adjusting the presentation of text in a Rapid Serial Visual Presentation (RSVP) display. Appx RSVP is a way of presenting text in a serial manner. Segments of text are sequentially presented on a display in a single row for a fixed period of time. Appx RSVP was one way of presenting information on a small device without increasing the size of the device and compromising portability. Appx , Appx In the mid s researchers explored the RSVP technique as a possible solution for displaying text on the small screens of mobile devices such as Personal Digital Assistants (PDAs). Appx Goldstein purports to take the RSVP process a step further, by controlling the text through user gaze inputs. Specifically, Goldstein describes a system that selectively adjusts presentation of text on an RSVP display by detecting the eye gaze direction of the reader. Appx , Appx Following detection of the change in point of gaze, text presentation is adjusted in a specific corresponding relationship to the detected change. Appx Figure 1 illustrates the different user gaze angles to be detected by Goldstein s invention on a 13

21 device: Appx2346. Within the mobile device 10 is a window 12 for displaying a text segment 14 on a single line. Appx The reader s points of gaze on the device are demonstrated by items 24 and 68. Appx

22 0021. The sensors (items 20 & 22) described by Goldstein are eye tracking sensors. Appx The only specific eye tracking sensor (or eye tracker) taught by Goldstein for use in its invention is the eye tracking device developed by the IBM Corporation at its Almaden Research Center, which is referred to by the acronym MAGIC. Id. This eye tracker device was disclosed in detail in a paper by Zhai et al. Appx Goldstein makes clear that while his invention is not limited to the eye tracker taught by Zhai et al., it is necessary to know whether the pupils of the user s eyes 18 are directed to a point of gaze 24, located within window 12 and thus focused upon text segments therein, or are directed to any location outside the window 12, such as to point of gaze 68. Appx Goldstein teaches that this eye gaze detection involves a geometric computation that provides an estimate of a reader s point of gaze in terms of coordinates on the display window 12. Appx In short, Goldstein only teaches eye tracking sensors which are used to detect[ ] change in a reader s point of gaze. Appx Goldstein does not mention eye contact sensors or eye contact sensing. Importantly, while Goldstein generically posits that its system is implemented using a mobile device, Goldstein does not provide 15

23 any detail regarding how to actually implement its teachings on such a mobile device. Appx With respect to mobile or cellular phones, specifically, Goldstein merely states that mobile phones are mobile devices that are increasingly used to acquire information, but Goldstein does not provide any discussion regarding how to implement its teachings on a mobile or cellular telephone. Appx III. INTER PARTES REVIEW In the IPRs at issue, the Board instituted review of Claim 14 of the 660 Patent and Claim 11 of the 665 Patent based on the grounds of anticipation by Goldstein under 35 U.S.C. 102(e) and obviousness under 35 U.S.C. 103(a) in view of Ho and Rochelle. 2 Appx349, Appx1233 A. Samsung s Initial Grounds of Unpatentability Claim 14 of the 660 Patent and Claim 11 of the 665 Patent each require a hardware sensor in or on [a cellular telephone] that senses attention of the user specifically toward the [cellular telephone]. Appx150 at 21:50-63, Appx150 at 22:28-29, Appx170 at 22:22-35, Appx170 at 23:5-6. In its Petitions, Samsung identified the eye tracking 2 In its final written decisions, the Board elected not to address this ground of unpatentability. Appx38, Appx101 16

24 sensors 20 and 22 disposed on a mobile communication device 10 as meeting this element. Appx , Appx , Appx , Appx That is, Goldstein s eye tracking sensor was specifically used by Samsung to teach the claimed hardware sensor, and Goldstein s mobile communication device was specifically used by Samsung to teach the claimed cellular telephone. Id. In particular, Samsung relied on the MAGIC eye tracker as taught by Zhai et al. Appx193, Appx1076, Appx Samsung was similarly clear that these eye tracking sensors of Goldstein determine[] the point of gaze or focus, with respect to the screen, of the pupils of the user's eyes. Appx194, Appx1077, Appx Samsung s Petitions did not identify any eye contact sensors or state that this element could be met by any sensor that performed automated pause/resume functionality other than the expressly disclosed eye tracking sensors. Compare Appx , Appx ( Goldstein teaches eye tracking sensors 20 and 22 the sensors 20 and 22 could, for example, comprise an eye tracking device Goldstein sensors 20 and 22 are hardware sensors ) with Appx , Appx ( Samsung relies on Goldstein s pause/resume functionality to anticipate the 660 patent claims. Goldstein s pause/resume functionality 17

25 requires detection of eye contact with the display. ). The Board s institution decisions contain little discussion of the hardware sensor, but do quote Samsung s argument that [b]y pausing and resuming presentation of text content based on point of gaze data provided by the hardware sensor 20/22 and a control signal representative of whether the point of gaze data indicates that the reader is looking at the display. Appx334, Appx B. Queen s University s Patent Owner Responses Following institution of the IPRs, Queen s University filed its Patent Owner Responses arguing that Goldstein was not enabled. Specifically, Queen s University argued that it was not within the skillset of a POSA, and a POSA would not have been able, to implement the eye-tracking system of Goldstein on a cell phone without undue experimentation. Appx , Appx Queen s University supported its arguments with declarations from Dr. Ravin Balakrishnan and Dr. Jeff Pelz. The Responses pointed out clear shortcomings of Goldstein with respect to implementing an eye-tracking sensor on a cellular phone. Samsung s Petitions relied on the MAGIC eye-tracker or other suitable eye-tracking devices as taught by Goldstein to meet the hardware 18

26 sensor limitation of the claims. The Petitions are, however, silent as to how one of ordinary skill would actually implement this sensor on a mobile device such as a cellular telephone. Indeed, Samsung s original expert, Dr. Turnbull, admitted that he did not consider enablement in reaching his conclusions and admitted he had little experience with the relevant hardware sensors. Appx at 238:25-239:7, Appx8883 at 97:8-24, Appx8922 at 136:4-20, Appx at 126:7-128:11. Queen s University also noted that the MAGIC tracker was not a stand-alone system, but rather used a desktop computer for implementation. Appx , Appx , Appx Eye-trackers at the time of the invention suffered from significant issues that prevented their use on mobile devices. Appx , Appx These issues include but are not limited to issues with operating systems, Appx5906, Appx5966, the size of the eye tracker relative to the mobile device, Appx449, Appx , Appx1333, Appx , Appx , Appx , Appx , processor speed and memory capacity, Appx468, Appx1352, Appx5906, Appx , Appx5843, Appx , and the fact that eye-trackers were reliant on individual positional calibration of gaze direction relative to the display, Appx , 19

27 Appx , Appx , Appx , Appx5966. As noted, as late as 2007, experts in the field continued to fail to implement eyetrackers on mobile devices. Appx469, Appx1353, Appx , Appx , Appx5966, Appx , Appx C. Samsung s New Grounds of Unpatentability and New Evidence in Support of Its Replies In reply to Queen s University s Responses, Samsung submitted 25- page Replies supplemented by a 68-page declaration from a new expert, Dr. Irfan Essa, relying on at least 65 other new pieces of evidence. Appx , Appx , Appx4692. Importantly, these documents did not actually address whether Goldstein s eye tracking sensors could have been implemented on a mobile device such as a cellular telephone. Rather, Samsung s Replies jettisoned its original theory that Goldstein taught and enabled eye tracking sensors replacing it with a new theory that eye contact detection could instead be used. Appx , Appx Not only did Samsung change the type of sensor it relied upon to teach the claimed sensor, but Samsung s new theory relied on a type of sensor not even disclosed by Goldstein. Id. Samsung now argued that it relied on Goldstein s pause/resume functionality to anticipate the 20

28 [ ] patent claims, id., and then looked elsewhere to see what kind of sensors could be used to implement that sole functionality, despite Goldstein s own disclosure that an eye tracking sensor would be used for that functionality (and others). Id. Specifically, Samsung argued that the pause/resume function could be implemented via eye contact detection on a mobile device. Id. Samsung even argued that Goldstein would have guided a POSITA to implement an eye sensor on a mobile device. Appx1542, Appx658. Samsung claimed that they did not need to show that the eye tracking sensors that they relied on in their Petitions were enabled (despite their Petitions relying on such sensors for anticipation (Appx , Appx )), but merely that pause/resume functionality was enabled by implement[ing] eye contact detection on a mobile device. Appx , Appx That is, to teach the sensor of the claims, Samsung now relied on a function of Goldstein, and then searched the art to find other sensors that could perform this function, despite Goldstein s explicit disclosure of an eye tracking sensor for that function. Indeed, Samsung s new expert even argued that a POSITA would have understood Goldstein as disclosing the use of eye contact sensors despite the fact that Goldstein mentions no such thing. Appx

29 In support of its new theory, Samsung presented substantial new evidence. This evidence included but was not limited to: (1) a paper by Morimoto that is not referenced by Goldstein; (2) evidence related to the Smart Bailando prototype including an undated paper by Goldstein that was not referenced in the Goldstein published application; (3) a description of the Tiqit eightythree handheld device; and (4) a detailed description of Dr. Essa s work with Myron Flickner. Appx , Appx , Appx , Appx , Appx , Appx None of this evidence was even hinted at in Samsung s Petitions. Importantly, Queen s University s challenge to the enablement by Goldstein of an eye tracking sensor on a mobile device was left entirely unanswered when Samsung raised its new theory in its Replies. This was a fundamental departure from the Petitions and not responsive to the Patent Owner Responses. Because of these deficiencies, Queen s University moved to strike the Replies, including the new 68-page declaration and 65 other exhibits. Appx , Appx In the alternative, Queen s University requested the ability to file surreplies with rebuttal evidence. Id. The Board refused to strike the Replies and new exhibits. Appx , Appx

30 1624. Instead, the Board authorized short, 10-page surreplies, but forbade Queen s University from providing any rebuttal evidence to address Samsung s newly disclosed evidence. Id. D. The Board s Reliance on Samsung s New Evidence and New Ground of Unpatentability, and Faulting Queen s University for not Presenting Evidence the Board Explicitly Forbade Given a posture where Samsung was allowed to present new, unrebutted evidence for the first time a reply, the Board found Claim 11 of the 665 Patent and Claim 14 of the 660 Patent anticipated. The Board expressly relied on the new ground of unpatentability that Goldstein s automated pause/resume functionality using an eye contact sensor was enabled to conclude that the claims were anticipated. Appx37, Appx100. In reaching its conclusion that Goldstein taught and enabled a hardware sensor in or on the device that senses attention of the user specifically toward the device, the Board relied almost exclusively on evidence that was presented for the first time in Samsung s Replies. Appx30-38, Appx The Board s conclusion is expressly premised on its agreement with Dr. Essa s conclusion, as summarized in paragraphs of his Declaration. Appx37, Appx100. The opinion describes and relies on the 23

31 Morimoto paper and the sensor technology taught therein. Appx30-38, Appx The Board considered the undated Goldstein paper for its teachings about the Smart Bailando prototype. 3 Appx33-37, Appx The Board also expressly relied on Dr. Essa s work with Myron Flickner. Appx30-38, Appx In order to show that the teachings of Goldstein could be implemented on a cellular device, the Board also relied on the previously undisclosed Tiqit eightythree handheld device. Appx32-36, Appx The Board gave little weight to Queen s University s expert because he failed to consider the evidence that was presented for the first time in Samsung s Replies despite the Board having denied Queen s University the opportunity to submit a declaration from the expert addressing that evidence. Appx32, Appx94. Specifically, the Board faulted Dr. Balakrishnan for not considering Morimoto or Dr. Essa s work with My- 3 The Board also relied on a Master s Thesis for its discussion of the Smart Bailando Prototype. See Appx33, Appx96. 24

32 ron Flicker, even though none of that evidence was presented by Samsung in its Petitions. 4 Id. Thus, the Board credited Dr. Essa s evidence that the Morimoto paper and the eye sensing described in Dr. Essa s work with Myron Flickner enabled Goldstein and found the claims invalid as anticipated. Id. (citations omitted). SUMMARY OF ARGUMENT The Board violated Queen s University s due process rights and the APA by relying on new grounds of unpatentability and new evidence while simultaneously refusing Queen s University any opportunity to present rebuttal evidence. The Board also erred in finding Goldstein enabled and anticipating. Consequently, the Board s determination that Claim 11 of the 665 Patent and Claim 14 of the 660 Patent are anticipated by Goldstein should be reversed. First, the Board erred as a matter of law by relying on new grounds of unpatentability and new evidence in its Final Written Decisions while 4 Samsung s expert at the time of the Petitions, on whose Declaration the Petitions rely, was Dr. Turnbull. It is telling that the Board s conclusions only minimally referenced Dr. Turnbull s testimony submitted with the Petitions, relying primarily on Dr. Essa s testimony submitted with the Petitioner s Replies. 25

33 denying Queen s University the chance to submit rebuttal evidence. Samsung did not identify in its Petitions any of the evidence of enablement of Goldstein that the Board ultimately relied on in its Final Written Decisions for the hardware sensor claim limitation. After Queen s University produced substantial evidence showing that the eye-tracking sensors taught by Goldstein were not enabled on a cellular phone as required by the claims, Samsung developed a new theory of anticipation and provided new evidence to support it. Throughout the IPR proceedings, Queen s University was denied any opportunity to rebut the evidence that supported the Board s ultimate basis for finding unpatentability. In relying on new grounds and new evidence for its determination of unpatentability, the Board exceeded its authority, violating the Administrative Procedure Act. Accordingly, the Board s determination of anticipation should be reversed. Second, the Board erred in finding the Goldstein reference enabled. The only hardware sensor taught by Goldstein is an eye-tracking sensor that requires the functionality of determining points of gaze, which requires calculating the estimated coordinates of a reader s point of gaze on the display window. In response to Queen s University s evidence that 26

34 such a sensor was not enabled on a cellular telephone, Samsung changed theories and argued that the pause/resume functionality of Goldstein could be enabled by eye contact sensors. Goldstein does not teach eye contact sensors. A prior art reference cannot be found anticipating when it teaches one thing and enables another. Accordingly, the Board s determination of anticipation should be reversed. Third, the Board erred in combining the teachings of multiple references to reach its conclusion of anticipation. As noted, Goldstein does not teach eye contact sensors, eye sensing, or even mere eye detection. Goldstein teaches eye-tracking and eye-gaze direction measurement to determine an x,y coordinate of gaze within a window on a display. To cure this deficiency in Goldstein, the Board examined the teachings of several other references and imported the teachings of those references related to eye detection and eye contact sensing to fill the gaps in Goldstein. Specifically, the Board replaced an explicit teaching in Goldstein (the eye tracking sensor) with another type of sensor (an eye contact sensor) allegedly found in the art that could have been used instead of the disclosed sensor. Appx32, Appx95. While such connecting the dots may be useful in evaluating obviousness, combinations that replace a reference s 27

35 explicit disclosure with alternatives that may have been found in the art is not appropriate for anticipation. Accordingly, the Board s determination of anticipation should be reversed. ARGUMENT I. STANDARDS OF REVIEW This Court reviews the Board s conclusions of law without deference and its findings of fact for substantial evidence. Microsoft Corp. v. Proxyconn, Inc., 789 F.3d 1292, 1297 (Fed. Cir. 2015) (citing In re Gartside, 2013 F.3d 1305, 1316 (Fed. Cir. 2000)); HP Inc. v. MPHJ Tech. Inv., LLC, 817 F.3d 1339, 1344 (Fed. Cir. 2016) ( anticipation is a question of fact reviewed for substantial evidence ) (internal citations omitted). The substantial evidence standard requires the reviewing court to ask whether a reasonable person might find that the evidentiary record supports the agency s conclusion. On-Line Careline, Inc. v. Am. Online, Inc., 229 F.3d 1080, 1085 (Fed. Cir. 2000). Enablement of a prior art reference is a question of law based on underlying factual findings. Antor, 689 F.3d at The ultimate question of enablement is reviewed without deference. Impax Labs., Inc. v. Aventis Pharm., Inc., 545 F.3d 1312, 1315 (Fed. Cir. 2008). 28

36 The petitioner in an IPR bears the burden of proving unpatentability. 35 U.S.C. 316(e). That burden, like the burden of production regarding invalidity issues, never shifts to the Patent Owner. See In re Magnum Oil Tools Int l, Ltd., 829 F.3d 1364, 1375 (Fed. Cir. 2016) The Court reviews the Board s procedures for compliance with the Administrative Procedure Act, 5 U.S.C. 551 et seq. See Dell Inc. v. Acceleron, LLC, 818 F.3d 1293, 1298 (Fed. Cir. 2016) (citing Belden Inc. v. Berk-Tek LLC, 805 F.3d 1064, 1080 (Fed. Cir. 2015). Under the Administrative Procedure Act, [the Court] must hold unlawful and set aside agency action... not in accordance with law [or]... without observance of procedure required by law. In re NuVasive, Inc., 841 F.3d 966, 970 (Fed. Cir. 2016) (quoting 5 U.S.C. 706). Whether the Board relied on a new ground of unpatentability in its final written decision in an IPR is a legal issue that is reviewed de novo on appeal. Id. II. THE BOARD ERRED BY RELYING ON NEWLY PRESENTED EVIDENCE AND NEW GROUNDS OF UNPATENTABILITY. The Board violated Queen s University s due process rights and the APA by relying on new grounds of unpatentability and new evidence that 29

37 was not present in Samsung s Petitions or the Board s Institution Decisions. The Board allowed Samsung to change its anticipation theory and present sixty-six new pieces of evidence including a 68-page declaration of a new expert for the first time in its Reply briefs. The Board erred by refusing to allow Queen s University the opportunity to present rebuttal evidence to Samsung s newly presented evidence. The Board compounded this error by relying exclusively on this new evidence in reaching its conclusion finding Goldstein enabled. A patent owner in Queen s University s position is undoubtedly entitled to notice of and a fair opportunity to meet the grounds of rejection. Belden, 805 F.3d at The indispensable ingredients of due process are notice and an opportunity to be heard by a disinterested decisionmaker. Id. (quoting Abbott Labs. v. Cordis Corp., 710 F.3d 1318, 1328 (Fed.Cir.2013)). For a formal adjudication like the inter partes review considered here, the APA imposes particular requirements on the PTO. Dell, 818 F.3d at First, the PTO must timely inform the patent owner of the matters of fact and law asserted. Id. (citing 5 U.S.C. 554(b)(3)). Second, the PTO must provide all interested parties opportunity for the submission and consideration of facts [and] arguments... 30

38 [and] hearing and decision on notice. Id. (citing 5 U.S.C. 554(c)) (brackets in original). Third, and important to the present case, the PTO must allow a party... to submit rebuttal evidence... as may be required for a full and true disclosure of the facts. Id. (citing 5 U.S.C. 556(d)) (emphasis added). Under the APA s standards, [a patent owner is] entitled to an adequate opportunity to respond to [newly asserted facts]. NuVasive, 841 F.3d at 972. This case is not the first time the Board violated the APA in refusing to give a patent owner appropriate leave to respond to new grounds of unpatentability or new evidence. For instance, in Dell, the Board erred in relying on a theory of anticipation that was raised for the first time at oral argument. Dell, 818 F.3d at In that case the disputed claim required caddies. Id. at In its petition, the petitioner only relied on the prior art reference s articulating door 262 as meeting the caddies limitation. Id. at In its reply, petitioner argued for the first time that the caddies required by claim 3 are found not only in [the reference s] articulating door 262 (previously argued) but also in [the references s] power-supply mounting mechanisms 278 (newly argued). Id. The patent owner sought the Board s permission to move to strike this 31

39 new argument or file a surreply. Id. The Board refused the request. Id. At oral argument, petitioner added yet another new argument, claiming that slides shown in Figure 12 of the reference also met the caddies limitation. Id. In finding the relevant claim invalid, the Board relied on the slides of Figure 12 as meeting the caddies limitation. Id.at This Court vacated the Board s ruling, holding that the Board denied [patent owner] notice and a fair opportunity to respond to this basis of cancellation. Id. at In reaching this holding, the Court reasoned that the patent owner had no opportunity to supply evidence, whether expert or lay or documentary evidence, about what the [] Figure 12 slides are whether they meet all the claim requirements. Id. Because the Board relied on that evidence for an essential part of its decision, reversal was necessary. Id. The procedural facts of Nuvasive are similar. In that case there were two IPRs at issue. The Court affirmed the Board s decision in one and reversed in the other. NuVasive, 841 F.3d at 968. Both petitions relied in part on a reference called Michelson. Id. at 969. In the affirmed decision, the petitioner expressly discussed Michelson s Figure 18 in its petition, but in the decision that was reversed, the petitioner did not. Id. 32

40 Following the patent owner s responses, the petitioner replied relying on Michelson s Figure 18. Id. Nuvasive objected to this argument as a new ground of invalidity and requested leave to file motions to strike or, alternatively, surreplies. Id. at 970. The Board denied both requests. Id. The Board also refused to allow Nuvasive to address the issue at oral argument. Id. In its written decisions in both IPRs the Board relied on the arguments regarding Figure 18. Id. at In the Board s decision that was reversed, the Court noted that because Figure 18 was not discussed in the petition, Nuvasive did not have the opportunity to respond to the Figure 18 arguments in its patent owner response. Id. at 972. In defending the Board s decisions, the PTO argued that Nuvasive was permitted to cross-examine [petitioner s expert] and to file observations on the cross-examination. Id. at 973. The Court stated that [o]bservations are not a vehicle for submitting new evidence, including new expert declarations, by the patent owner. Id. The Court held that the opportunity to file observations was not enough because they are not a substitute for the opportunity to present arguments and evidence. Id. 33

41 The procedural facts of the present case are even more egregious than those of Dell and Nuvasive. In each of those cases, the new grounds of invalidity and new evidence were found in a prior art reference disclosed in the opening petitions. Here, the Board relied on evidence and references that were not contained anywhere in the Petitions, but rather were presented by Samsung for the first time in its Replies. In its Petitions, Samsung relied on eye tracking sensors 20 and 22 to meet the hardware sensor element of the claims at issue. Appx , Appx , Appx , Appx In response, Queen s University presented compelling evidence that Goldstein failed to enable the use of eye tracking sensors on a cellular phone as the claims require. Appx , Appx Samsung then jettisoned its expert, Dr. Turnbull, who admittedly knew little about the relevant hardware. Appx8883 at 97:8-24, Appx8922 at 136:4-20, Appx at 126:7-128:11. For the Replies, Samsung found an entirely new expert, Dr. Essa, who provided a sixty-eight page declaration and sixty five new references. Samsung used this new declaration and evidence to argue that Goldstein enabled the use of eye contact detection on a mobile device. Appx , Appx Samsung even expressly disavowed its 34

42 need to show that the eye tracking sensors relied on in its Petitions were enabled, Appx , Appx , 5 and instead embraced a new argument that the pause/resume functionality of Goldstein was enabled by implement[ing] eye contact detection on a mobile device. Appx , Appx That is, Samsung jettisoned its link in the Petitions between Goldstein s eye tracking sensor and the claimed hardware sensor, Appx , Appx , and changed its argument to combine Goldstein with other references, substituting an eye contact sensor for Goldstein s disclosed eye tracking sensor. Appx , Appx This was not just showing that Goldstein s disclosed eye tracking sensor could have been implemented with an eye contact sensor (which, while false, would have at least been responsive to the enablement challenge). Rather, this constituted an entirely new obviousness-type argument, presented by a full substitution of a disclosed sensor in the primary reference for a different type of sensor allegedly found in the art that had not previously been relied upon. 5 In fact, Samsung went so far as to argue that its Petitions clearly relied upon Goldstein for disclosure of eye contact detection, despite the fact that the Petitions and Goldstein mention no such thing. Appx1616, Appx732 35

43 Samsung s Petitioner s Replies and supporting exhibits were thus not responsive to the Patent Owner Responses; they were an effort to rehabilitate the deficiencies of the Petitions. The declaration and supporting exhibits were not used to simply explain the disclosure of Goldstein, as may be appropriate under the case law. Teleflex, Inc. v. Ficosa N. Am. Corp., 299 F.3d 1313, 1335 (Fed. Cir. 2002) (citing In re Donohue, 766 F.2d at 534). A substitution of elements is not the same as simply informing how to implement the disclosed elements. It is not proper to identify an element in the prior art disclosure (i.e., the eye tracking sensor), isolate one function out of the many performed by that element (i.e., the pause/resume functionality), identify some other element from the art capable of performing the isolated function (i.e., an eye contact sensor), and swap out the disclosed element with the other element, under the guise of anticipation. Given this mountain of new evidence and arguments, Queen s University properly objected and filed a motion to strike, or in the alternative to file a surreply with supporting evidence. Appx , Appx The Board effectively denied Queen s University s motion, only allowing ten-page surreplies without evidence. Appx , Appx

44 1624. Without any ability to provide evidence, Queen s University was denied an adequate opportunity to rebut new evidence from Dr. Essa and Samsung. This denial of due process was protested at the oral hearing as well, in an extensive colloquy with the panel. Appx Nonetheless, the Board proceeded to rely on Samsung s new evidence in its written decisions even faulting Queen s University s expert for not providing rebuttal evidence, after the Board explicitly denied any opportunity to do so. Appx32, Appx94. Newly presented evidence relied on by the Board includes the Morimoto paper, the undated Goldstein reference, the Tiqit eightythree handheld device, and Dr. Essa s sixty-eight page declaration describing his work with Myron Flickner that was previously undisclosed by Samsung. Appx30-38, Appx The Board should have allowed Queen s University to present the evidence, including a new expert declaration addressing: (1) feasibility of implementing the system of Goldstein on the Tiqit eightythree handheld device; (2) Dr. Essa s assertion that a POSITA would have understood Goldstein as disclosing the use of eye contact sensors despite its express teachings to the contrary, which was not identified by Samsung in its 37

45 Petitions; and (3) Dr. Essa s contention that new undated Goldstein paper represented successful implementation of the claims. Queen s University requested this relief, Appx , Appx , but was denied. Appx , Appx The Board s refusal to allow Queen s University to file rebuttal evidence violated its due process rights under the APA. The APA requires the Board to allow parties the ability to submit rebuttal evidence. Dell, 818 F.3d at 1301 (citing 5 U.S.C. 556(d)). The Board s failure to do so in this case unquestionably prejudiced Queen s University because it allowed the evidence submitted by Samsung in its Replies and relied on by the Board to go unrebutted. The fact that the Board allowed Queen s University to submit short surreplies does not save the Board s decisions. A surreply, like the observations in Nuvasive, is not a vehicle for submitting new evidence, including new expert declarations, by the patent owner. NuVasive, 841 F.3d at 973. As such, a surreply, without supporting evidence, is not enough because it is not a substitute for the opportunity to present... evidence. Id. Because the Board violated the Queen s University s due process rights under the APA, the Board s decisions should be reversed. 38