DETAILED ACTION
This Office Action is in response to the remarks entered on 03/06/2026. Claims 1, 16, 18, 20 are amended. Claim 17 is cancelled. Claim 21 is newly added. Claims 1-16, 18-21 are pending.
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Arguments
The Applicant’s arguments regarding the rejection of above claims have been fully considered.
In reference to Applicant’s arguments about:
35 USC 101 rejections.
Examiner’s response:
Rejections are withdrawn in view of claim amendments and Applicant’s arguments.
In reference to Applicant’s arguments about:
35 USC 103 rejections.
Examiner’s response:
Applicant’s arguments have been fully considered but are moot in view of new grounds of rejections.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1-5, 10-16, 18-21 are rejected under 35 U.S.C. 103 as being unpatentable over Dally et al (US Pub. No. 2020/0249674- hereinafter Dally) in view of Gupta (US Pub. No. 2018/0196436- hereinafter Gupta).
Referring to Claim 1, Dally teaches a method, comprising:
obtaining a training dataset associated with an autonomous vehicle action, the training dataset comprising trajectory data for a plurality of examples of the autonomous vehicle action and labels for each of the plurality of examples (see Dally at [0013], wherein it describes the use of the invention towards autonomous vehicles: “Approaches in accordance with various embodiments provide for the navigation of controllable objects, such as autonomous vehicles or robots”. Further at [0050], it can be seen how the training process is achieved: “The request can then be directed to a training manager 622, which can select an appropriate model or network and then train the model using relevant training data 624”, and “The tree management component 628 can then generate a decision tree using sequences of possible actions and probable reactions, and generate scores for each sequence using a selected value function”, and “A path can be selected using the highest path score, and a next option provided to an optimizer 630, which can also be located on the vehicle in at least some embodiments. The optimizer 630, which can also be internal to the control system 606, can provide navigation actions that can be used to control the vehicle and cause the vehicle to proceed along the selected path”. Therefore, this training data described by Dally corresponds to the claimed ‘training dataset’, as it uses the mapping of a reaction based on an action (interpreted as labels) and paths for training autonomous vehicles to control the vehicle and cause the vehicle to proceed along a selected path);
generating a decision tree based on the trajectory data and the labels of the training dataset (see Dally at [0050]: “The tree management component 628 can then generate a decision tree using sequences of possible actions and probable reactions, and generate scores for each sequence using a selected value function” and “A path can be selected using the highest path score, and a next option provided to an optimizer 630, which can also be located on the vehicle in at least some embodiments. The optimizer 630, which can also be internal to the control system 606, can provide navigation actions that can be used to control the vehicle and cause the vehicle to proceed along the selected path”);
determining a vehicle trajectory standard including one or more rules for operating an autonomous vehicle based on the decision tree (see Dally at [0023]: “As discussed herein, a value function can then be utilized to determine a value for each path, or sequence of actions. The path leading to the highest value leaf node may then be selected as the five-to-ten second path for the vehicle” and “The function may also include rewards for making progress towards the destination, making a successful lane change, avoiding collisions, providing a smooth ride, and keeping safe distances, among other such options”. Further at [0050]: “The tree search and inference in many instances will run on the vehicle, and not on a separate system or in the cloud. A path can be selected using the highest path score, and a next option provided to an optimizer 630, which can also be located on the vehicle in at least some embodiments. The optimizer 630, which can also be internal to the control system 606, can provide navigation actions that can be used to control the vehicle and cause the vehicle to proceed along the selected path”. Therefore, the functions including rewards for making a successful lane, avoiding collisions and keeping safe distances are inherently analogous to the rules for operating an autonomous vehicle); and
communicating the vehicle trajectory standard to the autonomous vehicle, wherein the autonomous vehicle is configured to use the vehicle trajectory standard to generate a vehicle trajectory and operate according to the generated vehicle trajectory, wherein the generated vehicle trajectory satisfies the one or more rules included in the vehicle trajectory standard (see Dally at [0023]: “As discussed herein, a value function can then be utilized to determine a value for each path, or sequence of actions. The path leading to the highest value leaf node may then be selected as the five-to-ten second path for the vehicle” and “The function may also include rewards for making progress towards the destination, making a successful lane change, avoiding collisions, providing a smooth ride, and keeping safe distances, among other such options”. Further at [0050]: “The tree search and inference in many instances will run on the vehicle, and not on a separate system or in the cloud. A path can be selected using the highest path score, and a next option provided to an optimizer 630, which can also be located on the vehicle in at least some embodiments. The optimizer 630, which can also be internal to the control system 606, can provide navigation actions that can be used to control the vehicle and cause the vehicle to proceed along the selected path”. Therefore, the navigation actions used to control the vehicle and cause the vehicle to proceed along the selected path is interpreted as the claimed generation of vehicle trajectory and operation of it. Further, Dally teaches the functions including rewards for making a successful lane, avoiding collisions and keeping safe distances, which are inherently analogous to the trajectory satisfying the rules).
However, Dally fails to explicitly teach determining a vehicle trajectory standard including one or more rules for operating an autonomous vehicle based on the decision tree, and
generate a vehicle trajectory and operate according to the generated vehicle trajectory, wherein the generated vehicle trajectory satisfies the one or more rules included in the vehicle trajectory standard.
Gupta teaches, in an analogous system:
determining a vehicle trajectory standard including one or more rules for operating an autonomous vehicle based on the decision tree (see Gupta at [0051]: First, the process 600 determines whether an initial vehicle trajectory satisfies a set of soft constraints (decision 602). As described previously with regard to FIG. 3, a “soft constraint” is a non-detrimental restriction on the calculated driving route for the autonomous vehicle. Ideally, a soft constraint should not be breached by the autonomous vehicle during travel, but violating a soft constraint generally does not inflict immediate damage to, or destruction of, the vehicle. As soft constraint usually pertains to the “rules of the road”. A “hard constraint” or detrimental restriction on the calculated driving route for the autonomous vehicle. A hard constraint cannot be breached by the autonomous vehicle during travel without inflicting damage to the autonomous vehicle, including possible destruction of the autonomous vehicle. It is important to note a trajectory which satisfies the set of soft constraints automatically satisfies the set of hard constraints”. Therefore, these soft and hard constraints based on rules of the road for determining an autonomous vehicle trajectory is interpreted as the claimed one or more rules for operating an autonomous vehicles), and
generate a vehicle trajectory and operate according to the generated vehicle trajectory, wherein the generated vehicle trajectory satisfies the one or more rules included in the vehicle trajectory standard (see Gupta at [0029]: “The steering mechanism 116 is configured to autonomously maneuver the vehicle according to the optimized vehicle trajectory that is generated by the vehicle trajectory optimization module 114”. Further at [0051]: First, the process 600 determines whether an initial vehicle trajectory satisfies a set of soft constraints (decision 602). As described previously with regard to FIG. 3, a “soft constraint” is a non-detrimental restriction on the calculated driving route for the autonomous vehicle. Ideally, a soft constraint should not be breached by the autonomous vehicle during travel, but violating a soft constraint generally does not inflict immediate damage to, or destruction of, the vehicle. As soft constraint usually pertains to the “rules of the road”. A “hard constraint” or detrimental restriction on the calculated driving route for the autonomous vehicle. A hard constraint cannot be breached by the autonomous vehicle during travel without inflicting damage to the autonomous vehicle, including possible destruction of the autonomous vehicle. It is important to note a trajectory which satisfies the set of soft constraints automatically satisfies the set of hard constraints”. Therefore, these soft and hard constraints based on rules of the road for determining an autonomous vehicle trajectory for ultimately maneuvering the vehicle is interpreted as the claimed operation of the vehicle satisfying the one or more rules).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the teachings of Dally with the above teachings of Gupta by determining a vehicle trajectory standard to operate an autonomous vehicle, as taught by Dally, wherein the standard includes rules for ultimately operating the autonomous vehicle, as taught by Gupta. The modification would have been obvious because one of ordinary skill in the art would be motivated to generate a more accurate and optimal route for the autonomous vehicle to travel (as suggested by Gupta at [0031]: “The optimized autonomous vehicle trajectory 210 is an improved version of the potential autonomous vehicle trajectory 208 that considers detected soft constraints and hard constraints in order to generate a more accurate and optimal route for the autonomous vehicle to travel”).
Referring to Claim 2, the combination of Dally and Gupta teaches the method of claim 1, wherein obtaining the training dataset associated with the autonomous vehicle action comprises:
obtaining first trajectory data associated with a first example of the autonomous vehicle action (see Dally at [0018]: “Accordingly, approaches in accordance with various embodiments can attempt to include predictions of other actors or objects (e.g., vehicles, pedestrians, cyclists, etc.) in the path planning process. A sequence of possible actions for all actors can be considered and taken into account when attempting to determine the best action for the respective vehicle to take next. This can include, for example, using a tree search to predict different trajectories for the other actors in response to various actions that may be taken by the vehicle to be navigated”. These are interpreted as trajectory data examples);
obtaining a first label for the first example of the autonomous vehicle action (see Dally at [0018]: “Accordingly, approaches in accordance with various embodiments can attempt to include predictions of other actors or objects (e.g., vehicles, pedestrians, cyclists, etc.) in the path planning process. A sequence of possible actions for all actors can be considered and taken into account when attempting to determine the best action for the respective vehicle to take next. This can include, for example, using a tree search to predict different trajectories for the other actors in response to various actions that may be taken by the vehicle to be navigated”. Therefore, Dally describes that according to the scenario, a particular or best action is considered based on the tree, interpreted as a label according to an example); and
associating the first label with the first trajectory data (see Dally at [0018]: “Accordingly, approaches in accordance with various embodiments can attempt to include predictions of other actors or objects (e.g., vehicles, pedestrians, cyclists, etc.) in the path planning process. A sequence of possible actions for all actors can be considered and taken into account when attempting to determine the best action for the respective vehicle to take next. This can include, for example, using a tree search to predict different trajectories for the other actors in response to various actions that may be taken by the vehicle to be navigated”. Therefore, Dally describes that according to the scenario, a particular or best action is considered based on the tree, interpreted as the association. Also, as previously explained by Dally at [0050]: “A path can be selected using the highest path score, and a next option provided to an optimizer 630, which can also be located on the vehicle in at least some embodiments. The optimizer 630, which can also be internal to the control system 606, can provide navigation actions that can be used to control the vehicle and cause the vehicle to proceed along the selected path”).
Referring to Claim 3, the combination of Dally and Gupta teaches the method of claim 2, wherein obtaining the first label for the first example of the autonomous vehicle action comprises receiving an annotation from a user device, and wherein the annotation indicates a user experience of the first example of the autonomous vehicle action (see Dally at [0065]: “The user may specify a recipe that indicates which attributes and attribute transformations are available for model training. The user may also specify various training parameters that control certain properties of the training process and of the resulting model”. Therefore, these recipes/inputs by the user for training the model is interpreted as the claimed ‘user experience’).
Referring to Claim 4, the combination of Dally and Gupta teaches the method of claim 2, wherein obtaining the first label for the first example of the autonomous vehicle action comprises receiving an annotation from a simulation system, and wherein the simulation system determines the annotation based on the first trajectory data satisfying at least one safety threshold (see Dally at [0023]: “If path planning is viewed as a multi-player game with sequences of actions taken by each actor, where those actions may depend at least in part upon the actions of others, then the set of possible actions at each stage, point, step, or level can be used to generate a decision tree that includes all possible options for each vehicle” and “The function may also include rewards for making progress towards the destination, making a successful lane change, avoiding collisions, providing a smooth ride, and keeping safe distances, among other such options. Various different value functions can be used that consider different value criteria, and there may be different weightings applied to different value criteria depending upon the current situation”. Therefore, since the path planning is viewed by Dally as a multiplayer game, this is interpreted as ‘simulation system’, and since Dally weights different criteria and one of them is maintaining a safe distance, this is interpreted as the safety threshold).
Referring to Claim 5, the combination of Dally and Gupta teaches the method of claim 1, wherein generating the decision tree based on the trajectory data and the labels of the training dataset comprises:
receiving the trajectory data and the labels at a root node of the decision tree, the trajectory data comprising a set of traces that each correspond to a label and an example (see Dally at [0013]: “Machine learning can be used to determine the probabilities, as well as to project out the options along the branches and paths of the decision tree including the sequences”. Further, at [0028]: “A root node 302 represents the current placement of the vehicles, and may reference other information as well, such as current speed or acceleration, etc. The sequence of actions is considered as a turn-based game, where nodes 304 at a first level each correspond to actions that can be taken by the present vehicle (shaded). This can include, for example, paths for up to the nine possible movement actions (AR, MS, etc.)”);
determining a first condition from a set of conditions that satisfies a branching condition (see Dally at [0024]: “At each level of the decision tree, a given node may then have nine branches, each corresponding to one of the potential motion options. As discussed herein, a probability may be determined for each of those options, which can be factors into the value determination for a given branch”);
branching the decision tree to a first node (see Dally at [0024]: “At each level of the decision tree, a given node may then have nine branches, each corresponding to one of the potential motion options. As discussed herein, a probability may be determined for each of those options, which can be factors into the value determination for a given branch”);
associating the root node with the first condition and a first sub-set of traces of the set of traces that satisfy the first condition with the first node (see Dally at [0013]: “Machine learning can be used to determine the probabilities, as well as to project out the options along the branches and paths of the decision tree including the sequences. In some embodiments only actions with at least a minimum probability are considered. In another embodiment, actions can be considered based on factors such as the corresponding amount of risk or loss, favorability, occupant comfort, and the like. A value function can be used to generate a value for each considered sequence, or path, and a proposed navigation path having a highest value can be selected”);
determining a continuation condition is satisfied (see Dally at [0013]: “Machine learning can be used to determine the probabilities, as well as to project out the options along the branches and paths of the decision tree including the sequences. In some embodiments only actions with at least a minimum probability are considered. In another embodiment, actions can be considered based on factors such as the corresponding amount of risk or loss, favorability, occupant comfort, and the like. A value function can be used to generate a value for each considered sequence, or path, and a proposed navigation path having a highest value can be selected”); and
in response to determining the continuation condition is satisfied, recursively, until the continuation condition is not satisfied, determining a second condition that satisfies the branching condition, branching the decision tree to second node, and associating the first node with the second condition and a second sub-set of the first sub-set of traces that satisfy the second condition with the second node (see Dally at [0026]: “Thus, all options involving a right turn may be excluded from consideration, eliminating those branches from the decision tree. Further, since the present car 202 is just ahead and to the right of that car 204, and the car 204 has a goal of avoiding collisions, the car 204 may have a very low probability of accelerating and turning to the left at the next point in time. Thus, this path option may be excluded from consideration (at least for path planning purposes) as well”. Further, at [0028]: “The sequence of actions is considered as a turn-based game, where nodes 304 at a first level each correspond to actions that can be taken by the present vehicle (shaded). This can include, for example, paths for up to the nine possible movement actions (AR, MS, etc.)).
Referring to Claim 10, the combination of Dally and Gupta teaches the method of claim 5, wherein the set of conditions comprise different types of conditions, and wherein each condition comprises at least one conditional operator and at least one variable (see Dally at [0013]: “Machine learning can be used to determine the probabilities, as well as to project out the options along the branches and paths of the decision tree including the sequences. In some embodiments only actions with at least a minimum probability are considered. In another embodiment, actions can be considered based on factors such as the corresponding amount of risk or loss, favorability, occupant comfort, and the like. A value function can be used to generate a value for each considered sequence, or path, and a proposed navigation path having a highest value can be selected”. Therefore, the actions for the path to take vary according to the risk or loss, favorability, occupant comfort, and the like, being the conditions).
Referring to Claim 11, the combination of Dally and Gupta teaches the method of claim 10, wherein the at least one variable is adjusted based on the set of traces (see Dally at [0013]: “Machine learning can be used to determine the probabilities, as well as to project out the options along the branches and paths of the decision tree including the sequences. In some embodiments only actions with at least a minimum probability are considered. In another embodiment, actions can be considered based on factors such as the corresponding amount of risk or loss, favorability, occupant comfort, and the like. A value function can be used to generate a value for each considered sequence, or path, and a proposed navigation path having a highest value can be selected”. Therefore, the value function that generates a value for each considered sequence corresponds to the variable adjusted based on the traces).
Referring to Claim 12, the combination of Dally and Gupta teaches the method of claim 5, wherein the continuation condition is satisfied when a sub-set of the set of traces that satisfy the second condition contains only positively labeled traces (see Dally at [0035]: “The “move generator” DNN can generate the most likely move for non-critical actors, and up to three or four most probable moves for critical actors (such as the actors directly adjacent the present vehicle)”, and “A positive value can be associated with achieving the respective goal, such as by successfully moving into the right lane”).
Referring to Claim 13, the combination of Dally and Gupta teaches the method of claim 5, wherein determining the vehicle trajectory standard based on the decision tree comprises: determining the vehicle trajectory standard by traversing nodes of the decision tree and joining conditions associated with traversed nodes (see Dally at [0040]: “The value of a state (and hence of the move to that state) can be calculated in some embodiments by traversing the tree to a chosen depth, such as to level ten. At the leaves of the tree, the future contribution to the value function can be estimated using a “value network” that takes the current state and the goal as input and returns a value”).
Referring to Claim 14, the combination of Dally and Gupta teaches the method of claim 1, wherein the decision tree comprises at least two levels and at least one condition at each level of the at least two levels (see Dally at [0029]: “The action for the present vehicle can be determined in response to that possible action by the other vehicle as a branch to a node 308 at the next level. This process can continue with a number of levels corresponding to the time scale in some embodiments, such as out to five or ten seconds with each level corresponding to a 0.25 second increment in one embodiment. The nodes of the last level can then correspond to leaf nodes at the end of the various paths, where the value determinations for the paths can be made. It should be understood that leaf nodes may exist at other levels as well, such as where a vehicle might reach a destination, collision, or other endpoint along a given path”. Therefore, Dally mentions several levels, interpreted as at least two levels).
Referring to Claim 15, the combination of Dally and Gupta teaches the method of claim 14, wherein each level of the at least two levels sorts the trajectory data into different groups in accordance with the at least one condition at each level (see Dally at [0029]: “The action for the present vehicle can be determined in response to that possible action by the other vehicle as a branch to a node 308 at the next level. This process can continue with a number of levels corresponding to the time scale in some embodiments, such as out to five or ten seconds with each level corresponding to a 0.25 second increment in one embodiment. The nodes of the last level can then correspond to leaf nodes at the end of the various paths, where the value determinations for the paths can be made. It should be understood that leaf nodes may exist at other levels as well, such as where a vehicle might reach a destination, collision, or other endpoint along a given path”).
Referring to Claim 16, the combination of Dally and Gupta teaches the method of claim 1, wherein the autonomous vehicle uses the vehicle trajectory standard to generate a plurality of trajectories, the method further comprising selecting an initial trajectory from the plurality of trajectories, wherein the initial trajectory satisfies the vehicle trajectory standard (see Dally at [0013]: “Machine learning can be used to determine the probabilities, as well as to project out the options along the branches and paths of the decision tree including the sequences. In some embodiments only actions with at least a minimum probability are considered. In another embodiment, actions can be considered based on factors such as the corresponding amount of risk or loss, favorability, occupant comfort, and the like”. Further at [0050]: “The tree search and inference in many instances will run on the vehicle, and not on a separate system or in the cloud. A path can be selected using the highest path score, and a next option provided to an optimizer 630, which can also be located on the vehicle in at least some embodiments. The optimizer 630, which can also be internal to the control system 606, can provide navigation actions that can be used to control the vehicle and cause the vehicle to proceed along the selected path”. Therefore, the options are interpreted as the plurality of trajectories, and the highest path score option is interpreted as the initial trajectory used as it satisfies the vehicle trajectory standard. Further, as explained at Claim 1, the combination of Dally and Gupta explains the standards as satisfying rules).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the teachings of Dally with the above teachings of Gupta by determining a vehicle trajectory standard to operate an autonomous vehicle, as taught by Dally, wherein the standard includes rules, as taught by Gupta. The modification would have been obvious because one of ordinary skill in the art would be motivated to generate a more accurate and optimal route for the autonomous vehicle to travel (as suggested by Gupta at [0031]: “The optimized autonomous vehicle trajectory 210 is an improved version of the potential autonomous vehicle trajectory 208 that considers detected soft constraints and hard constraints in order to generate a more accurate and optimal route for the autonomous vehicle to travel”).
Referring to independent Claim 18 and Claim 20, they are rejected on the same basis as independent claim 1 since they are analogous claims.
Referring to dependent Claim 19, it is rejected on the same basis as dependent claim 5 since they are analogous claims.
Referring to Claim 21, the combination of Dally and Gupta teaches the method of claim 1, wherein the one or more rules include one or more legal requirements for operating a vehicle (see Dally at [0039]: “As mentioned, the present vehicle path determination system or manager can select the move or action that maximizes an expected value function” and “A legality term can result in varying negative values being applied for breaking laws, with the magnitude depending on the law. For example, running a red light would likely come with a much more significant penalty than exceeding the speed limit by 1 mph. The penalties could be chosen such that a vehicle is enabled to break a law (or at least certain laws) when needed to avoid a collision. The law able to be broken may also depend upon the type of collision or object of the collision in some embodiments”. Furthermore, Gupta teaches at [0051]: “As soft constraint usually pertains to the “rules of the road””, and at [0057]: “In some embodiments, road data may also include additional attributes of the road, such as a number of lanes, speed limits, lane marking types, road bank and slope angles, or the like”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the teachings of Dally with the above teachings of Gupta by determining a vehicle trajectory standard to operate an autonomous vehicle, as taught by Dally, wherein the standard includes rules, as taught by Gupta. The modification would have been obvious because one of ordinary skill in the art would be motivated to generate a more accurate and optimal route for the autonomous vehicle to travel (as suggested by Gupta at [0031]: “The optimized autonomous vehicle trajectory 210 is an improved version of the potential autonomous vehicle trajectory 208 that considers detected soft constraints and hard constraints in order to generate a more accurate and optimal route for the autonomous vehicle to travel”).
Claims 6-9 are rejected under 35 U.S.C. 103 as being unpatentable over Dally in view of Gupta and further in view of Bergmeir (NPL “Classifying component failures of a vehicle fleet”- hereinafter Bergmeir).
Referring to Claim 6, the combination of Dally and Gupta teaches the method of claim 5, however, fails to teach wherein the branching condition is satisfied when the first condition has a highest impurity reduction measure.
Bergmeir teaches, in an analogous system, wherein the branching condition is satisfied when the first condition has a highest impurity reduction measure (see Bergmeir at p. 34 second paragraph: “Rather the “optimal” pair of variable and split point depends on a so called impurity measure [50] that is used to compute the cutpoint values” and third paragraph: “Thereby, a variable and a cutpoint are considered as being optimal for a split if they lead to the highest empirical impurity reduction, e.g., the maximal decrease of the Gini Index [50] or the highest Information Gain [64]”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the combination of Dally and Gupta with the above teachings of Bergmeir by determining a first condition from a set of conditions that satisfies a branching condition in a decision tree, as taught by the combination of Dally and Gupta, when the first condition has a highest impurity reduction measure, as taught by Bergmeir. The modification would have been obvious because one of ordinary skill in the art would be motivated to reach an optimal split using the highest empirical impurity reduction measure.
Referring to Claim 7, the combination of Dally and Gupta teaches the method of claim 5, however, fails to teach wherein determining the first condition from the set of conditions that satisfies the branching condition comprises: determining an impurity reduction measure for each condition of the set of conditions; and selecting the first condition based on the first condition having a highest impurity reduction measure.
Bergmeir teaches, in an analogous system, wherein determining the first condition from the set of conditions that satisfies the branching condition comprises: determining an impurity reduction measure for each condition of the set of conditions; and selecting the first condition based on the first condition having a highest impurity reduction measure (see Bergmeir at p. 34 second paragraph: “Rather the “optimal” pair of variable and split point depends on a so called impurity measure [50] that is used to compute the cutpoint values” and third paragraph: “Thereby, a variable and a cutpoint are considered as being optimal for a split if they lead to the highest empirical impurity reduction, e.g., the maximal decrease of the Gini Index [50] or the highest Information Gain [64]”, and fourth paragraph: “
PNG
media_image1.png
74
480
media_image1.png
Greyscale
. “).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the combination of Dally and Gupta with the above teachings of Bergmeir by determining a first condition from a set of conditions that satisfies a branching condition in a decision tree, as taught by the combination of Dally and Gupta, when the first condition has a highest impurity reduction measure, as taught by Bergmeir. The modification would have been obvious because one of ordinary skill in the art would be motivated to reach an optimal split using the highest empirical impurity reduction measure.
Referring to Claim 8, the combination of Dally, Gupta and Bergmeir teaches the method of claim 7, wherein determining the impurity reduction measure for each condition of the set of conditions comprises, for each condition: determining a first set of traces that satisfy the condition and a second set of traces that do not satisfy the condition; and determining the impurity reduction measure based on the first set of traces and the second set of traces (see Bergmeir at p. 34 third paragraph: “Thereby, a variable and a cutpoint are considered as being optimal for a split if they lead to the highest empirical impurity reduction, e.g., the maximal decrease of the Gini Index [50] or the highest Information Gain [64]”. Therefore, the Gini impurity by definition is the difference between the traces being classified into a correct class (or satisfying a condition), therefore, if one trace satisfy a condition (e.g. yes) and another trace does not satisfy a condition (e.g. no), it means the split is impure (as evidenced by Fatih Karabiber in NPL: “Giny Impurity” at page 1 second paragraph “To find the best feature for the first split of the tree - the root node - you could calculate how poorly each feature divided the data into the correct class, default “yes”) or didn't default “no”). This calculation would measure the impurity of the split, and the feature with the lowest impurity would determine the best feature for splitting the current node”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the combination of Dally and Gupta with the above teachings of Bergmeir by determining a first condition from a set of conditions that satisfies a branching condition in a decision tree, as taught by the combination of Dally and Gupta, when the first condition has a highest impurity reduction measure, as taught by Bergmeir. The modification would have been obvious because one of ordinary skill in the art would be motivated to reach an optimal split using the highest empirical impurity reduction measure.
Referring to Claim 9, the combination of Dally, Gupta and Bergmeir teaches the method of claim 7, wherein the impurity reduction measure is at least one of an information gain, a Gini gain, or a misclassification gain (see Bergmeir at p. 34 second paragraph: “Rather the “optimal” pair of variable and split point depends on a so called impurity measure [50] that is used to compute the cutpoint values” and third paragraph: “Thereby, a variable and a cutpoint are considered as being optimal for a split if they lead to the highest empirical impurity reduction, e.g., the maximal decrease of the Gini Index [50] or the highest Information Gain [64]”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the combination of Dally and Gupta with the above teachings of Bergmeir by determining a first condition from a set of conditions that satisfies a branching condition in a decision tree, as taught by the combination of Dally and Gupta, when the first condition has a highest impurity reduction measure, as taught by Bergmeir. The modification would have been obvious because one of ordinary skill in the art would be motivated to reach an optimal split using the highest empirical impurity reduction measure.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to LUIS A SITIRICHE whose telephone number is (571)270-1316. The examiner can normally be reached M-F 9am-6pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, David Yi can be reached at (571) 270-7519. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/LUIS A SITIRICHE/Primary Examiner, Art Unit 2126
Prosecution Insights