CN113392566A - Simulation optimization design method based on difference - Google Patents
Simulation optimization design method based on difference Download PDFInfo
- Publication number
- CN113392566A CN113392566A CN202110788921.9A CN202110788921A CN113392566A CN 113392566 A CN113392566 A CN 113392566A CN 202110788921 A CN202110788921 A CN 202110788921A CN 113392566 A CN113392566 A CN 113392566A
- Authority
- CN
- China
- Prior art keywords
- population
- individual
- sub
- vector
- design method
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000005457 optimization Methods 0.000 title claims abstract description 39
- 238000013461 design Methods 0.000 title claims abstract description 36
- 238000000034 method Methods 0.000 title claims abstract description 29
- 238000004088 simulation Methods 0.000 title claims abstract description 21
- 230000007246 mechanism Effects 0.000 claims abstract description 9
- 239000011159 matrix material Substances 0.000 claims abstract description 5
- 230000035772 mutation Effects 0.000 claims description 28
- 238000012360 testing method Methods 0.000 claims description 13
- 230000003044 adaptive effect Effects 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000009826 distribution Methods 0.000 claims description 4
- 239000000126 substance Substances 0.000 claims description 3
- 230000008844 regulatory mechanism Effects 0.000 abstract description 2
- 238000004904 shortening Methods 0.000 abstract 1
- 238000005516 engineering process Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/23—Design optimisation, verification or simulation using finite element methods [FEM] or finite difference methods [FDM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a simulation optimization design method based on difference, and relates to the field of EDA tool simulation optimization. The invention introduces a multi-population strategy cooperative mechanism so as to achieve the purpose of multi-physics cooperative optimization. Firstly, setting parameters to be optimized and an optimization target of a chip, initializing a population, adopting a variation strategy of a multi-population mechanism to ensure the individual cooperativity in the evolution process, then properly rotating a target individual and the variation individual by an initialized coordinate system through a covariance learning matrix in the evolution process, using a mixed intersection strategy to achieve better global optimization, simultaneously adopting a self-adaptive parameter regulation mechanism to enhance the diversity and the high efficiency of the population, improving the global optimization capability, realizing the automatic design and optimization of the parameters in the EDA tool simulation, effectively shortening the design cycle of an engineering product, and providing possibility for more effectively solving the problem of modern engineering design.
Description
Technical Field
The invention relates to the field of EDA tool simulation optimization, in particular to a simulation optimization design method based on difference.
Background
The Integrated Circuit (IC) industry is the core of the information technology industry, being a strategic, fundamental and precedent industry that supports economic social development and ensures national security. An autonomous high-end, reliable, safe and controllable Electronic Design Automation (EDA) software tool is an important digital infrastructure for developing the integrated circuit industry. Currently, the IC design industry is developing towards the direction of high integration, super large scale, high performance, low power consumption, and in addition, the huge challenge brought to the design by the nano-scale advanced process, powerful EDA tools can help the design engineer to solve various potential problems, improve the reliability of the chip, shorten the design cycle, accelerate the mass production of the chip, and improve the market competitiveness of the product. However, as the application of the EDA technology in the field of electronic information is increased, some deep-level design service problems are faced, a complex product design task flow often needs to use a plurality of different EDA design software tools, and after a plurality of steps are completed cooperatively, design files and data formats used by design tool software of different manufacturers are compatible, and cannot be shared, so that a fully-automatic design flow cannot be formed.
Disclosure of Invention
The technical problem to be solved by the invention is to provide a simulation optimization design method based on difference, which realizes the rapid optimization of parameters, improves the global optimization capability, shortens the design period of engineering products and can more effectively improve the automation degree of modern engineering design.
The invention provides a simulation optimization design method based on difference, which comprises the following steps:
step 1, setting parameters to be optimized and an optimization target of a chip, initializing a population, wherein the parameters comprise a population scale NP, an individual dimension D and an evolution maximum iteration number TMAXCurrent iteration times t;
step 2, calculating the adaptive value f (x) of population individualsi,t);
Step 3, dividing the whole population P into three sub-populations P1, P2 and P3, wherein the size of the P1 population is larger than that of the P2 and P3;
4, performing mutation operation cooperatively through a plurality of mutation strategies;
step 5, calculating the excellent rate of each generation of sub-populationkAccording to the rate of the next generation evolution initialization stagekNewly assigning the sub-populations for each variation strategy;
step 6, updating the target individual and the variant individual through a characteristic-based cooperative system;
step 7, performing mixed crossing by adopting a mixed crossing strategy;
step 8, selecting a mechanism: calculating test individualsFitness value ofIf it isReplacing the current target individual with the test individual; otherwise, keeping the current target individual;
step 9, recombining the three sub-populations into an overall population, and recording the current optimal individual of the population;
step 10, judging whether a termination condition t is met>TMAXAnd if so, outputting the optimal solution, otherwise, returning to the step 2.
Further, the population is initialized in step 1 by using the following formula:
wherein, i is 1,2, is, NP, T is 1,2, TMAXJ ═ 1,2, ·, D; NP represents the number of population individuals; d represents the dimension of the solution; wherein the content of the first and second substances,vector representing a feasible solution in the problem search space and representable as a D-dimensiont is the number of iterations,is the maximum value of the search space and,is the minimum value of the search space, and rand (0, 1) represents a range of [0,1 ]]Uniformly distributed random decimal fractions are fit for.
Further, the step 3 of dividing the whole population P into three sub-populations P1, P2 and P3 is performed by using the following formula:
wherein, PkDenotes the kth sub-population, NPkDenotes the population size, σ, of the kth sub-populationkThe scale proportion of the kth sub-population is shown, and the scale proportion relation is sigma1>σ2=σ3And sigmak∈[0,1]。
Further, the scaling factor F is calculated as follows:
wherein formula N (0.5,0.15) represents a Gaussian distribution, f (u)i,t) Represents the test vector ui,tFitness of (a), f (x)i,t) Representing a target individual xi,tIs within the range of 0,1];Is a global optimal individual; r is1,r2,r3Is the number of three individuals randomly selected from the population and satisfies i ≠ r1 ≠ r2 ≠ r3, respectively represent a quilt r1,r2,r3Three individuals select a vector for performing mutation operation, also called a target vector, and a mutation vector obtained by differential mutation operationReferred to as the contribution vector to individual i.
Further, the plurality of mutation strategies include "best/1", "current-to-rand/1", and "rand/2":
wherein F is a scaling factor, and K is in the range of 0,1],Is a global optimal individual; r is1,r2,r3Is the number of three individuals randomly selected from the population and satisfies i ≠ r1 ≠ r2 ≠ r3,respectively represent a quilt r1,r2,r3Three individuals select a vector for performing mutation operation, also called a target vector, and a mutation vector obtained by differential mutation operationReferred to as the contribution vector to individual i.
Further, the sub-population yield in step 5 is calculated as follows:
counting the number of good individual dimensions of the sub-population which are reserved after each population evolution as bestnumkObtaining the sub-population excellent rate according to the dimension D of the population individualkThe sub-population yield is expressed by the formula
Wherein, NPkThe population size of the kth sub-population is shown.
Further, the hybrid crossing strategy in step 7 specifically includes:
where θ represents the broad value of the mixed crossover probability and the test vectorCalculated by a binomial intersection method, intersected according to the following formula and rotated back to the original coordinate system:
further, the step 6 further comprises: calculating covariance C of the sub-population through the sub-population information, solving an eigenvector matrix R of the covariance, and performing updating operation by adopting the following formula:
the invention has the advantages that:
by taking the minimization of some parameters of the chip, such as the highest temperature value of the radiating fin in the chip, as an optimization target, the optimal solution is obtained by carrying out the optimization design on the structural parameters of the radiating fin of the chip according to the method, the performance of the radiating fin of the chip can be improved, and the optimization of the design of a chip product is realized.
The population structure of a multi-population mechanism is adopted, the individual cooperativity in the evolution process is guaranteed by combining each sub-population with a corresponding variation strategy, a properly rotating coordinate system is established for cross operation through the covariance learning among the populations, and the global search capability is improved through adopting a mixed cross strategy, so that the accuracy of data is improved, and the automation capability of the EDA is improved.
Drawings
The invention will be further described with reference to the following examples with reference to the accompanying drawings.
FIG. 1 is an execution flow chart of a simulation optimization design method based on difference according to the present invention.
Detailed Description
As shown in fig. 1, the simulation optimization design method based on difference of the present invention includes:
step 1, setting parameters to be optimized and an optimization target of a chip, initializing a population, wherein the parameters comprise a population scale NP, an individual dimension D and an evolution maximum iteration number TMAXCurrent iteration times t; preferably, the population is initialized in step 1 by using the following formula:
wherein, i is 1,2, is, NP, T is 1,2, TMAXJ ═ 1,2, ·, D; NP represents the number of population individuals; d represents the dimension of the solution; wherein the content of the first and second substances,vector representing a feasible solution in the problem search space and representable as a D-dimensiont is the number of iterations,is the maximum value of the search space and,is the minimum value of the search space, and rand (0, 1) represents a range of [0,1 ]]Uniformly distributed random decimal fractions are fit for.
Step 2, calculating the adaptive value f (x) of population individualsi,t) (ii) a The calculation formula of the adaptive value is as follows:
f(xi,t)=0.1/(0.1+1/|lgf(xi,t)|),0≤f(xi,t)≤10-λ
in the formula, λ is determined by the calculation accuracy of the computer, and in one embodiment, λ is 8.
Step 3, dividing the whole population P into three sub-populations P1, P2 and P3, wherein the size of the P1 population is larger than that of the P2 and P3; preferably, the step 3 of dividing the whole population P into three sub-populations P1, P2 and P3 is performed by using the following formula:
wherein, PkDenotes the kth sub-population, NPkDenotes the population size, σ, of the kth sub-populationkThe scale proportion of the kth sub-population is shown, and the scale proportion relation is sigma1>σ2=σ3And sigmak∈[0,1]。
Step 4, performing mutation operation cooperatively through multiple mutation strategies, specifically, each population corresponds to one mutation strategy, and the multiple mutation strategies include "best/1", "current-to-rand/1", and "rand/2":
wherein F is a scaling factor, and K is in the range of 0,1],Is a global optimal individual; r is1,r2,r3Is the number of three individuals randomly selected from the population and satisfies i ≠ r1 ≠ r2 ≠ r3,are respectively provided withIs represented by1,r2,r3Three individuals select a vector for performing mutation operation, also called a target vector, and a mutation vector obtained by differential mutation operationReferred to as the contribution vector to individual i; wherein, the calculation formula of the scaling factor F is as follows:
wherein formula N (0.5,0.15) represents a Gaussian distribution, f (u)i,t) Represents the test vector ui,tFitness of (a), f (x)i,t) Representing a target individual xi,tIs within the range of 0,1];Is a global optimal individual; r is1,r2,r3Is the number of three individuals randomly selected from the population and satisfies i ≠ r1 ≠ r2 ≠ r3, respectively represent a quilt r1,r2,r3Three individuals select a vector for performing mutation operation, also called a target vector, and a mutation vector obtained by differential mutation operationReferred to as the contribution vector to individual i; if the current fitness is better, the parameter F in the last iteration is continuously reservedi,tOtherwise, a gaussian distribution N (0.5,0.15) is satisfied, with an expected value of 0.5 and a variance of 0.15.
Step 5, calculating the excellent rate of each generation of sub-populationkAccording to the rate of the next generation evolution initialization stagekRe-dividing into individual variation strategies (i.e.The three mutation strategies "best/1", "current-to-rand/1" and "rand/2") distribute the sub-populations; preferably, the sub-population yield in step 5 is calculated as follows: counting the number of good individual dimensions of the sub-population which are reserved after each population evolution as bestnumkObtaining the sub-population excellent rate according to the dimension D of the population individualkThe sub-population yield is expressed by the formula
Wherein, NPkThe population size of the kth sub-population is shown.
Step 6, updating the target individual and the variant individual through a characteristic-based cooperative system; preferably, the step 6 further comprises: calculating covariance C of the sub-population through the sub-population information, solving an eigenvector matrix R of the covariance, and performing updating operation by adopting the following formula:
step 7, performing mixed crossing by adopting a mixed crossing strategy; preferably, the hybrid crossing strategy in step 7 specifically includes:
where θ represents the broad value of the mixed crossover probability and the test vectorCalculated by a binomial intersection method, intersected according to the following formula and rotated back to the original coordinate system:
step 8, selecting a mechanism: calculating test individualsFitness value ofIf it isReplacing the current target individual with the test individual; otherwise, keeping the current target individual;
step 9, recombining the three sub-populations into an overall population, and recording the current optimal individual of the population;
step 10, judging whether a termination condition t is met>TMAXAnd if so, outputting the optimal solution, otherwise, returning to the step 2.
The variant individuals are individuals subjected to variant operation, the target individuals are individuals selected by a selection mechanism, and the test individuals are individuals subjected to crossing.
The simulation optimization design method based on the difference introduces a multi-population strategy cooperative mechanism so as to achieve the purpose of multi-physical-field cooperative optimization. Firstly, a variation strategy of a multi-population mechanism is adopted to ensure the individual cooperativity in the evolution process, then, in the evolution process, an initialized coordinate system is properly rotated with a target individual and the variation individual through a covariance learning matrix, the global search capability is improved by using a mixed intersection strategy to achieve better global optimization, meanwhile, a self-adaptive parameter regulation mechanism is adopted to enhance the diversity and the high efficiency of the population to a certain extent and improve the overall optimization performance, and finally the obtained solution is applied to the parameter optimization of the product in the EDA tool (such as the optimization of parameters of temperature, voltage and the like of a chip product), so that the design accuracy and the efficiency of the EDA tool can be improved.
Although specific embodiments of the invention have been described above, it will be understood by those skilled in the art that the specific embodiments described are illustrative only and are not limiting upon the scope of the invention, and that equivalent modifications and variations can be made by those skilled in the art without departing from the spirit of the invention, which is to be limited only by the appended claims.
Claims (9)
1. A simulation optimization design method based on difference is characterized in that: the method comprises the following steps:
step 1, setting parameters to be optimized and an optimization target of a chip, initializing a population, wherein the parameters comprise a population scale NP, an individual dimension D and an evolution maximum iteration number TMAXCurrent iteration times t;
step 2, calculating the adaptive value f (x) of population individualsi,t);
Step 3, dividing the whole population P into three sub-populations P1, P2 and P3, wherein the size of the P1 population is larger than that of the P2 and P3;
4, performing mutation operation cooperatively through a plurality of mutation strategies;
step 5, calculating the excellent rate of each generation of sub-populationkAccording to the rate of the next generation evolution initialization stagekNewly assigning the sub-populations for each variation strategy;
step 6, updating the target individual and the variant individual through a characteristic-based cooperative system;
step 7, performing mixed crossing by adopting a mixed crossing strategy;
step 8, selecting a mechanism: calculating test individualsFitness value ofIf it isReplacing the current target individual with the test individual; otherwise, keeping the current target individual;
step 9, recombining the three sub-populations into an overall population, and recording the current optimal individual of the population;
step 10,Judging whether a termination condition t is met>TMAXAnd if so, outputting the optimal solution, otherwise, returning to the step 2.
2. The differential-based simulation optimization design method of claim 1, wherein: in step 1, the population is initialized by using the following formula:
wherein, i is 1,2, is, NP, T is 1,2, TMAXJ ═ 1,2, ·, D; NP represents the number of population individuals; d represents the dimension of the solution; wherein the content of the first and second substances,vector representing a feasible solution in the problem search space and representable as a D-dimensiont is the number of iterations,is the maximum value of the search space and,is the minimum value of the search space, and rand (0, 1) represents a range of [0,1 ]]Uniformly distributed random decimal fractions are fit for.
3. The differential-based simulation optimization design method of claim 1, wherein: in the step 3, the step of dividing the whole population P into three sub-populations P1, P2 and P3 is carried out by adopting the following formula:
wherein,PkDenotes the kth sub-population, NPkDenotes the population size, σ, of the kth sub-populationkThe scale proportion of the kth sub-population is shown, and the scale proportion relation is sigma1>σ2=σ3And sigmak∈[0,1]。
4. The differential-based simulation optimization design method of claim 1, wherein: the scaling factor F is calculated as follows:
wherein formula N (0.5,0.15) represents a Gaussian distribution, f (u)i,t) Represents the test vector ui,tFitness of (a), f (x)i,t) Representing a target individual xi,tIs within the range of 0,1];Is a global optimal individual; r is1,r2,r3Is the number of three individuals randomly selected from the population and satisfies i ≠ r1 ≠ r2 ≠ r3, respectively represent a quilt r1,r2,r3Three individuals select a vector for performing mutation operation, also called a target vector, and a mutation vector obtained by differential mutation operationReferred to as the contribution vector to individual i.
5. The differential-based simulation optimization design method of claim 1, wherein: the multiple variation strategies include "best/1", "current-to-rand/1", and "rand/2":
wherein F is a scaling factor, and K is in the range of 0,1],Is a global optimal individual; r is1,r2,r3Is the number of three individuals randomly selected from the population and satisfies i ≠ r1 ≠ r2 ≠ r3,respectively represent a quilt r1,r2,r3Three individuals select a vector for performing mutation operation, also called a target vector, and a mutation vector obtained by differential mutation operationReferred to as the contribution vector to individual i.
6. The differential-based simulation optimization design method of claim 1, wherein: the calculation method of the sub-population excellent rate in the step 5 is as follows:
counting the number of good individual dimensions of the sub-population which are reserved after each population evolution as bestnumkObtaining the sub-population excellent rate according to the dimension D of the population individualkThe sub-population yield is expressed by the formula
Wherein, NPkThe population size of the kth sub-population is shown.
7. The differential-based simulation optimization design method of claim 1, wherein: the hybrid crossing strategy in the step 7 specifically comprises:
where θ represents the broad value of the mixed crossover probability and the test vectorCalculated by a binomial intersection method, intersected according to the following formula and rotated back to the original coordinate system:
8. the differential-based simulation optimization design method of claim 1, wherein: the step 6 further comprises: calculating covariance C of the sub-population through the sub-population information, solving an eigenvector matrix R of the covariance, and performing updating operation by adopting the following formula:
9. the differential-based simulation optimization design method of claim 1, wherein: the chip parameter in the step 1 is the temperature of the heat sink in the chip, and the optimization target is the minimum of the highest temperature value of the heat sink in the chip.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110788921.9A CN113392566A (en) | 2021-07-13 | 2021-07-13 | Simulation optimization design method based on difference |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110788921.9A CN113392566A (en) | 2021-07-13 | 2021-07-13 | Simulation optimization design method based on difference |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113392566A true CN113392566A (en) | 2021-09-14 |
Family
ID=77626142
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110788921.9A Pending CN113392566A (en) | 2021-07-13 | 2021-07-13 | Simulation optimization design method based on difference |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113392566A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117422003A (en) * | 2023-12-19 | 2024-01-19 | 深圳市德兰明海新能源股份有限公司 | Method and device for optimally designing radiating fin and storage medium |
CN117910410B (en) * | 2024-03-19 | 2024-05-31 | 电子科技大学 | Large-scale multi-target simulation chip circuit evolution optimization design method |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1205863A1 (en) * | 2000-11-14 | 2002-05-15 | Honda R&D Europe (Deutschland) GmbH | Multi-objective optimization |
CN108564592A (en) * | 2018-03-05 | 2018-09-21 | 华侨大学 | Based on a variety of image partition methods for being clustered to differential evolution algorithm of dynamic |
CN109002877A (en) * | 2018-04-18 | 2018-12-14 | 华侨大学 | Multiple dimensioned collaboration differential evolution optimization method |
-
2021
- 2021-07-13 CN CN202110788921.9A patent/CN113392566A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1205863A1 (en) * | 2000-11-14 | 2002-05-15 | Honda R&D Europe (Deutschland) GmbH | Multi-objective optimization |
CN108564592A (en) * | 2018-03-05 | 2018-09-21 | 华侨大学 | Based on a variety of image partition methods for being clustered to differential evolution algorithm of dynamic |
CN109002877A (en) * | 2018-04-18 | 2018-12-14 | 华侨大学 | Multiple dimensioned collaboration differential evolution optimization method |
Non-Patent Citations (2)
Title |
---|
杜永兆 等: "多种群协方差学习差分进化算法", 《电子信息学报》, 30 June 2019 (2019-06-30), pages 1488 - 1495 * |
梁静等: "基于混合策略的差分进化算法", 《郑州大学学报(工学版)》, vol. 34, no. 05, 30 September 2013 (2013-09-30), pages 59 - 62 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117422003A (en) * | 2023-12-19 | 2024-01-19 | 深圳市德兰明海新能源股份有限公司 | Method and device for optimally designing radiating fin and storage medium |
CN117422003B (en) * | 2023-12-19 | 2024-05-17 | 深圳市德兰明海新能源股份有限公司 | Method and device for optimally designing radiating fin and storage medium |
CN117910410B (en) * | 2024-03-19 | 2024-05-31 | 电子科技大学 | Large-scale multi-target simulation chip circuit evolution optimization design method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Jin et al. | A systems approach to evolutionary multiobjective structural optimization and beyond | |
US10089421B2 (en) | Information processing apparatus and information processing method | |
García-Nieto et al. | Parallel multi-swarm optimizer for gene selection in DNA microarrays | |
Marcoulides et al. | Specification searches in structural equation modeling with a genetic algorithm | |
CN108983722B (en) | Optimized scheduling method for final test of integrated circuit chip | |
CN113392566A (en) | Simulation optimization design method based on difference | |
Kureichik et al. | Placement of VLSI fragments based on a multilayered approach | |
Pan et al. | Solving the sampling problem of the sycamore quantum supremacy circuits | |
CN112418431A (en) | Method and system for mixing models | |
CN111554346B (en) | Protein sequence design implementation method based on multi-objective optimization | |
Sun et al. | Protein function prediction using function associations in protein–protein interaction network | |
CN107220463A (en) | One kind mixing polarity XNOR/OR circuit area optimization methods | |
CN113222160B (en) | Quantum state conversion method and device | |
CN114511094B (en) | Quantum algorithm optimization method and device, storage medium and electronic device | |
US20220277120A1 (en) | Systems and methods for machine learning based fast static thermal solver | |
CN116306849A (en) | Training of reverse neural network model and determining method and device of optical processor | |
CN116090568A (en) | Method and device for determining size relation between quantum data and classical floating point data | |
CN115331754A (en) | Molecule classification method based on Hash algorithm | |
Termritthikun et al. | Evolutionary neural architecture search based on efficient CNN models population for image classification | |
US20210264081A1 (en) | Methods of designing semiconductor devices, design systems performing the same and methods of manufacturing semiconductor devices using the same | |
Schneider et al. | GPU-accelerated time simulation of systems with adaptive voltage and frequency scaling | |
TW202240455A (en) | Poly-bit cells | |
CN114512194A (en) | Method and device for acquiring target system test state in quantum chemical simulation | |
CN113362898A (en) | RNA subcellular localization method for identifying by fusing multiple sequence frequency information | |
Yan et al. | Discriminant space metric network for few-shot image classification |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210914 |
|
RJ01 | Rejection of invention patent application after publication |