DETAILED ACTION
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA . This action is responsive to the Application filed on 02/05/2024. Claims 1-12 are pending in the case. Claims 1, 6, and 8 are independent claims.
Claim Rejections - 35 U.S.C. § 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 of this title, 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.
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 are advised of the obligation under 37 C.F.R. § 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, 6, and 8 are rejected under 35 U.S.C. § 103 as being unpatentable over Schütt et al. (Schütt, Kristof, Oliver Unke, and Michael Gastegger. "Equivariant message passing for the prediction of tensorial properties and molecular spectra." In International conference on machine learning, pp. 9377-9388. PMLR, 2021, hereinafter Schuett) in view of Kearnes et al. (Kearnes, Steven, Kevin McCloskey, Marc Berndl, Vijay Pande, and Patrick Riley. "Molecular graph convolutions: moving beyond fingerprints." Journal of computer-aided molecular design 30, no. 8 (2016): 595-608, hereinafter Kearnes).
As to independent claim 1, Schuett teaches
A method for a trained neural network, the method comprising (Title and abstract):
acquiring a plurality of inputs comprising a plurality of initial parameters for a polyatomic system (Figure 1, two structures);
passing the plurality of inputs through one or more rotationally equivariant message passing layers of the trained neural network to generate a plurality of outputs comprising an update to each of the plurality of inputs (Figure 1, "message passing using angles and directions for two structures. All edges within the cutoff range (dashed lines) have equal length. The representations of the blue and red node are the same using angles (left), while directions allow to distinguish both structures (right)"); and
determining a potential energy of the polyatomic system based on the plurality of outputs (Page 4, section 4, "potential energy surface E(Z1; : : : ;ZN; ~r1; : : : ; ~rN), with nuclear charges Zi 2 N and atom positions ~ri 2 R3" et seq.),….
Schuett does not appear to expressly teach each of the one or more rotationally equivariant message passing layers is constructed from one or more symmetric message functions.
Kearnes teaches each of the one or more rotationally equivariant message passing layers is constructed from one or more symmetric message functions (Page 598, equation 5 and figure 3).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, having the equivariant message passing neural network of Schuett to include the molecular machine learning techniques of Kearnes to allow the model to take greater advantage of information in the graph structure (see Kearnes at abstract).
As to independent claim 6, Schuett teaches
A system, comprising: a memory storing a trained geometric message passing neural network configured to predict potential energies and forces of atomic systems, the trained geometric message passing neural network comprising one or more rotationally equivariant message passing layers constructed from one or more symmetric message functions; and a processor configured with instructions in non-transitory memory that when executed cause the processor to (Title and abstract. Section 3, "Equivariant message passing"):…
pass the plurality of initial parameters through the one or more rotationally equivariant message passing layers to predict each of a potential energy and a plurality of forces of the atomic system of interest (Page 2, "the rotationally equivariant representation of PAINN enables the prediction of tensorial properties which we apply to the ring-polymer MD simulation of infrared and Raman spectra". Page 4, section 4, "potential energy surface E(Z1; : : : ;ZN; ~r1; : : : ; ~rN), with nuclear charges Zi 2 N and atom positions ~ri 2 R3" et seq.).
Schuett does not appear to expressly teach receive a plurality of initial parameters of an atomic system of interest.
Kearnes teaches receive a plurality of initial parameters of an atomic system of interest (Page 598, equation 5 and figure 3).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, having the equivariant message passing neural network of Schuett to include the molecular machine learning techniques of Kearnes to allow the model to take greater advantage of information in the graph structure (see Kearnes at abstract).
As to independent claim 8, Schuett teaches
A method for a neural network, the method comprising (Title and abstract):
training the neural network to predict each of a potential energy and a plurality of forces of an atomic system by (Page 6, "apply exponential smoothing with factor 0:9 to the validation loss to reduce the impact of fluctuations which are particularly common when training with both energies and forces"):
acquiring a set of training data comprising a plurality of samples of the atomic system (Page 6, "with 1k known structures of which we use 950 for training");…
penalizing deviations of the potential energy and the plurality of forces by minimizing a loss function with respect to a plurality of trainable parameters (Page 6, "this property both the squared loss as well as the MAE can be reduced when minimizing the MAE directly");…
predicting each of the potential energy and the plurality of forces of the atomic system by updating the plurality of initial parameters with the trained neural network (Page 2, "the rotationally equivariant representation of PAINN enables the prediction of tensorial properties which we apply to the ring-polymer MD simulation of infrared and Raman spectra". Page 4, section 4, "potential energy surface E(Z1; : : : ;ZN; ~r1; : : : ; ~rN), with nuclear charges Zi 2 N and atom positions ~ri 2 R3" et seq.).
Schuett does not appear to expressly teach passing the set of training data through one or more rotationally equivariant message passing layers constructed from one or more symmetric message functions; and receiving a plurality of initial parameters of the atomic system.
Kearnes teaches passing the set of training data through one or more rotationally equivariant message passing layers constructed from one or more symmetric message functions; and receiving a plurality of initial parameters of the atomic system (Page 598, equation 5 and figure 3).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, having the equivariant message passing neural network of Schuett to include the molecular machine learning techniques of Kearnes to allow the model to take greater advantage of information in the graph structure (see Kearnes at abstract).
Claim 2 is rejected under 35 U.S.C. § 103 as being unpatentable over Schuett in view of Kearnes and Gasteiger et al. (Gasteiger, Johannes, Janek Groß, and Stephan Günnemann. "Directional message passing for molecular graphs." arXiv preprint arXiv:2003.03123 (2020), hereinafter Gasteiger).
As to dependent claim 2, the rejection of claim 1 is incorporated.
Schuett does not appear to expressly teach each of the one or more symmetric message functions includes a radial Bessel function, a polynomial cutoff function, and a pair of multilayer perceptrons.
Gasteiger teaches each of the one or more symmetric message functions includes a radial Bessel function, a polynomial cutoff function, and a pair of multilayer perceptrons (Page 2, "the distance and angle can be jointly represented in a principled and effective manner by using spherical Bessel functions and spherical harmonics. We leverage these innovations to construct the directional message passing neural network (DimeNet). DimeNet can learn both molecular properties and atomic forces").
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, having the equivariant message passing neural network of Schuett to include the directional message passing for molecular graphs techniques of Gasteiger to achieve better performance than the currently prevalent Gaussian radial basis representations while using fewer than 1/4 of the parameters (see Gasteiger at abstract).
Claims 3 and 4 are rejected under 35 U.S.C. § 103 as being unpatentable over Schuett in view of Kearnes and Park et al. (Park, Cheol Woo, Mordechai Kornbluth, Jonathan Vandermause, Chris Wolverton, Boris Kozinsky, and Jonathan P. Mailoa. "Accurate and scalable graph neural network force field and molecular dynamics with direct force architecture." npj Computational Materials 7, no. 1 (2021): 73, hereinafter Park).
As to dependent claim 3, the rejection of claim 1 is incorporated.
Schuett does not appear to expressly teach the plurality of initial parameters comprises each of a plurality of atomic feature arrays, a plurality of latent force vectors, a plurality of total force vectors, a plurality of interatomic force vectors, a plurality of displacement vectors, and a plurality of interatomic distances for the polyatomic system.
Park teaches the plurality of initial parameters comprises each of a plurality of atomic feature arrays, a plurality of latent force vectors, a plurality of total force vectors, a plurality of interatomic force vectors, a plurality of displacement vectors, and a plurality of interatomic distances for the polyatomic system (Figure 1, "the node and edge embeddings respectively contain the atom type and interatomic distance information. The embeddings are then iteratively updated during the message passing stage. The final updated edge embeddings are used for predicting the interatomic force magnitudes. The force on the center atom j is calculated by summing the force contributions of neighboring atoms i 2 Nj that are calculated by multiplying the force magnitude and the respective unit vector." Page 3, "Mathematically the individual force contribution and the total force prediction can respectively be written as").
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, having the equivariant message passing neural network of Schuett to include the graph neural network techniques of Park to achieve high performance in terms of force prediction accuracy and computational speed (see Park at abstract).
As to dependent claim 4, Schuett further teaches the plurality of atomic feature arrays is rotationally invariant (Page 2, "Rotational invariance of the representation can be ensured by choosing rotationally invariant message and update functions").
Claims 5, 9, and 12 are rejected under 35 U.S.C. § 103 as being unpatentable over Schuett in view of Kearnes and Birchfield et al. (U.S. Pat. App. Pub. 2020/0301510, hereinafter Birchfield).
As to dependent claim 5, the rejection of claim 1 is incorporated.
Schuett does not appear to expressly teach One or more tangible, non-transitory storage medium storing executable instructions that, when executed by a processor, cause the processor to perform the method of claim 1.
Birchfield teaches One or more tangible, non-transitory storage medium storing executable instructions that, when executed by a processor, cause the processor to perform the method of claim 1 (Paragraph 113).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, having the equivariant message passing neural network of Schuett to include the force estimation using deep learning techniques of Birchfield to estimate force on a tactile sensor in a way that is accurate in both magnitude and direction, over a wide range of forces (see Birchfield at paragraph 1).
.
As to dependent claim 9, the rejection of claim 8 is incorporated.
Schuett does not appear to expressly teach minimizing the loss function with respect to the plurality of trainable parameters comprises minimizing an average cosine distance between a plurality of latent force vectors of the atomic system and a plurality of normalized reference force vectors of the atomic system.
Birchfield teaches minimizing the loss function with respect to the plurality of trainable parameters comprises minimizing an average cosine distance between a plurality of latent force vectors of the atomic system and a plurality of normalized reference force vectors of the atomic system (Paragraph 61, "directional error is computed as the cosine similarity between the vectors of the predicted force on the ground truth force").
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, having the equivariant message passing neural network of Schuett to include the force estimation using deep learning techniques of Birchfield to estimate force on a tactile sensor in a way that is accurate in both magnitude and direction, over a wide range of forces (see Birchfield at paragraph 1).
As to dependent claim 12, the rejection of claim 8 is incorporated.
Schuett does not appear to expressly teach One or more tangible, non-transitory storage medium storing executable instructions that, when executed by a processor, cause the processor to perform the method of claim 8.
Birchfield teaches One or more tangible, non-transitory storage medium storing executable instructions that, when executed by a processor, cause the processor to perform the method of claim 8 (Paragraph 113).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, having the equivariant message passing neural network of Schuett to include the force estimation using deep learning techniques of Birchfield to estimate force on a tactile sensor in a way that is accurate in both magnitude and direction, over a wide range of forces (see Birchfield at paragraph 1).
Claim 7 is rejected under 35 U.S.C. § 103 as being unpatentable over Schuett in view of Kearnes, Park, and Birchfield.
As to dependent claim 7, the rejection of claim 6 is incorporated.
Schuett does not appear to expressly teach predicting each of the potential energy and the plurality of forces of the atomic system of interest comprises: generating a plurality of latent force vectors based on Newton's third law.
Park teaches predicting each of the potential energy and the plurality of forces of the atomic system of interest comprises: generating a plurality of latent force vectors based on Newton's third law (Page 3, "For a GNNFF with L message-passing layers, the final updated state of edge eij, represented by embedding hL ði; jÞ, is used to calculate nij, a scalar quantity that denotes the magnitude of the force contribution that atom i is exerting on atom j. The force contribution of atom i onto j is given by niju ! ij and the total force prediction on atom j is simply the vector sum of the force contributions of all neighboring atoms.").
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, having the equivariant message passing neural network of Schuett to include the graph neural network techniques of Park to achieve high performance in terms of force prediction accuracy and computational speed (see Park at abstract).
Schuett does not appear to expressly teach minimizing an average cosine distance between the plurality of latent force vectors and a plurality of ground-truth force vectors of the atomic system of interest.
Birchfield teaches minimizing an average cosine distance between the plurality of latent force vectors and a plurality of ground-truth force vectors of the atomic system of interest (Paragraph 61, "directional error is computed as the cosine similarity between the vectors of the predicted force on the ground truth force").
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, having the equivariant message passing neural network of Schuett to include the force estimation using deep learning techniques of Birchfield to estimate force on a tactile sensor in a way that is accurate in both magnitude and direction, over a wide range of forces (see Birchfield at paragraph 1).
Claims 10 and 11 are rejected under 35 U.S.C. § 103 as being unpatentable over Schuett in view of Kearnes and Batzner et al. (Batzner, Simon, Albert Musaelian, Lixin Sun, Mario Geiger, Jonathan P. Mailoa, Mordechai Kornbluth, Nicola Molinari, Tess E. Smidt, and Boris Kozinsky. "E (3)-equivariant graph neural networks for data-efficient and accurate interatomic potentials." Nature communications 13, no. 1 (2022): 2453, hereinafter Batzner).
As to dependent claim 10, the rejection of claim 8 is incorporated.
Schuett does not appear to expressly teach a size of the set of training data is 1-10% of a size of a set of training data used to train other message passing neural networks and achieve comparable accuracy.
Batzner teaches a size of the set of training data is 1-10% of a size of a set of training data used to train other message passing neural networks and achieve comparable accuracy (Page 2, "NequIP exhibits exceptional data efficiency, enabling the construction of accurate interatomic potentials from limited data sets of fewer than 1000 or even as little as 100 reference ab-initio calculations, where other methods require orders of magnitude more").
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, having the equivariant message passing neural network of Schuett to include the graph neural network techniques of Batzner to allow for the construction of accurate potentials using high-order quantum chemical level of theory as reference and enables high-fidelity molecular dynamics simulations over long time scales (see Batzner at abstract).
As to dependent claim 11, Schuett further teaches the trained neural network predicts the potential energy and the plurality of forces faster than the other message passing neural networks for a given computing device implementing the trained neural network (Page 9, "equivariant message passing allows us to significantly reduce both model size and inference time compared to directional message-passing while retaining accuracy").
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Casey R. Garner whose telephone number is 571-272-2467. The examiner can normally be reached Monday to Friday, 8am to 5pm, Eastern Time.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Alexey Shmatov can be reached on 571-270-3428. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from Patent Center and the Private Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from Patent Center or Private PAIR. Status information for unpublished applications is available through Patent Center and Private PAIR to authorized users only. Should you have questions about access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free).
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) Form at https://www.uspto.gov/patents/uspto-automated- interview-request-air-form.
/Casey R. Garner/Primary Examiner, Art Unit 2123
Prosecution Insights