Pub Date : 2022-07-18DOI: 10.1109/CEC55065.2022.9870283
Hayden Andersen, Andrew Lensen, Will N. Browne, Yi Mei
Counterfactual explanations are a popular eXplainable AI technique, used to provide contrastive answers to “what-if” questions. These explanations are consistent with the way that an everyday person will explain an event, and have been shown to satisfy the ‘right to explanation’ of the European data regulations. Despite this, current work to generate counterfactual explanations either makes assumptions about the model being explained or utlises algorithms that perform suboptimally on continuous data. This work presents two novel algorithms to generate counterfactual explanations using Particle Swarm Optimization (PSO) and Differential Evolution (DE). These are shown to provide effective post-hoc explanations that make no assumptions about the underlying model or data structure. In particular, PSO is shown to generate counterfactual explanations that utilise significantly fewer features to generate sparser explanations when compared to previous related work.
{"title":"Evolving Counterfactual Explanations with Particle Swarm Optimization and Differential Evolution","authors":"Hayden Andersen, Andrew Lensen, Will N. Browne, Yi Mei","doi":"10.1109/CEC55065.2022.9870283","DOIUrl":"https://doi.org/10.1109/CEC55065.2022.9870283","url":null,"abstract":"Counterfactual explanations are a popular eXplainable AI technique, used to provide contrastive answers to “what-if” questions. These explanations are consistent with the way that an everyday person will explain an event, and have been shown to satisfy the ‘right to explanation’ of the European data regulations. Despite this, current work to generate counterfactual explanations either makes assumptions about the model being explained or utlises algorithms that perform suboptimally on continuous data. This work presents two novel algorithms to generate counterfactual explanations using Particle Swarm Optimization (PSO) and Differential Evolution (DE). These are shown to provide effective post-hoc explanations that make no assumptions about the underlying model or data structure. In particular, PSO is shown to generate counterfactual explanations that utilise significantly fewer features to generate sparser explanations when compared to previous related work.","PeriodicalId":153241,"journal":{"name":"2022 IEEE Congress on Evolutionary Computation (CEC)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128640734","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-18DOI: 10.1109/CEC55065.2022.9870380
Marina de la Cruz López, C. Cervigón, J. Alvarado, M. Botella, J. Hidalgo
People with diabetes have to properly manage their blood glucose levels in order to avoid acute complications. This is a difficult task and an accurate and timely prediction may be of vital importance, specially of extreme values. Perhaps one of the main concerns of people with diabetes is to suffer an hypoglycemia (low value) event and moreover, that the event will be prolonged in time. It is crucial to predict events of hyperglycemia (high value) and hypoglycemia that may cause health damages in the short term and potential permanent damages in the long term. The aim of this paper is to describe our research on predicting hypoglycemia events using Dynamic structured Grammatical Evolution. Our proposal gives white box models induced by a grammar based on if-then-else conditions. We trained and tested our system with real data collected from 5 different diabetic patients, producing 30 minutes predictions with encouraging results.
{"title":"Evolving Classification Rules for Predicting Hypoglycemia Events","authors":"Marina de la Cruz López, C. Cervigón, J. Alvarado, M. Botella, J. Hidalgo","doi":"10.1109/CEC55065.2022.9870380","DOIUrl":"https://doi.org/10.1109/CEC55065.2022.9870380","url":null,"abstract":"People with diabetes have to properly manage their blood glucose levels in order to avoid acute complications. This is a difficult task and an accurate and timely prediction may be of vital importance, specially of extreme values. Perhaps one of the main concerns of people with diabetes is to suffer an hypoglycemia (low value) event and moreover, that the event will be prolonged in time. It is crucial to predict events of hyperglycemia (high value) and hypoglycemia that may cause health damages in the short term and potential permanent damages in the long term. The aim of this paper is to describe our research on predicting hypoglycemia events using Dynamic structured Grammatical Evolution. Our proposal gives white box models induced by a grammar based on if-then-else conditions. We trained and tested our system with real data collected from 5 different diabetic patients, producing 30 minutes predictions with encouraging results.","PeriodicalId":153241,"journal":{"name":"2022 IEEE Congress on Evolutionary Computation (CEC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128656682","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-18DOI: 10.1109/CEC55065.2022.9870237
Jonatan Gómez, Elizabeth León Guzman
This paper analyzes bitstring functions, character-izes genes according to their contribution to the genome's fitness, and proposes an optimization algorithm (G ABO) that uses this characterization for directing the optimization process. We define a gene's contribution as the difference between the genome's fitness when the gene takes a value of 1 and its fitness when the gene takes a value of 0. We characterize a gene as intron-like if it does not contribute to the genome's fitness (zero difference) and as separable-like if its contribution to the fitness of both the genome and genome's complement is the same. Gabo divides genes into two groups coding-like and intron-like genes. Then it searches for an optimal solution by reducing intron-like genes (IOSA) and analyzing coding-like genes (COSA). G Aborepeats these two steps while there are intron-like genes, not all genes are separable-like, and function evaluations are available. We test the performance of Gabo on well-known binary-encoding functions and a function that we define as the mix of them. Our results indicate that G Aboproduces the optimal or near to the optimal solution on the tested functions expending a reduced number of function evaluations and outperforming well-established optimization algorithms.
{"title":"Gabo: Gene Analysis Bitstring Optimization","authors":"Jonatan Gómez, Elizabeth León Guzman","doi":"10.1109/CEC55065.2022.9870237","DOIUrl":"https://doi.org/10.1109/CEC55065.2022.9870237","url":null,"abstract":"This paper analyzes bitstring functions, character-izes genes according to their contribution to the genome's fitness, and proposes an optimization algorithm (G ABO) that uses this characterization for directing the optimization process. We define a gene's contribution as the difference between the genome's fitness when the gene takes a value of 1 and its fitness when the gene takes a value of 0. We characterize a gene as intron-like if it does not contribute to the genome's fitness (zero difference) and as separable-like if its contribution to the fitness of both the genome and genome's complement is the same. Gabo divides genes into two groups coding-like and intron-like genes. Then it searches for an optimal solution by reducing intron-like genes (IOSA) and analyzing coding-like genes (COSA). G Aborepeats these two steps while there are intron-like genes, not all genes are separable-like, and function evaluations are available. We test the performance of Gabo on well-known binary-encoding functions and a function that we define as the mix of them. Our results indicate that G Aboproduces the optimal or near to the optimal solution on the tested functions expending a reduced number of function evaluations and outperforming well-established optimization algorithms.","PeriodicalId":153241,"journal":{"name":"2022 IEEE Congress on Evolutionary Computation (CEC)","volume":"364 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130960377","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-18DOI: 10.1109/CEC55065.2022.9870275
Jose M. Tapia-Avitia, J. M. Cruz-Duarte, I. Amaya, J. C. Ortíz-Bayliss, H. Terashima-Marín, N. Pillay
Metaheuristics (MHs) are proven powerful algorithms for solving non-linear optimisation problems over discrete, continuous, or mixed domains. Applications have ranged from basic sciences to applied technologies. Nowadays, the literature contains plenty of MHs based on exceptional ideas, but often, they are just recombining elements from other techniques. An alternative approach is to follow a standard model that customises population-based MHs, utilising simple heuristics extracted from well-known MHs. Different approaches have explored the combination of such simple heuristics, generating excellent results compared to the generic MHs. Nevertheless, they present limitations due to the nature of the metaheuristic used to study the heuristic space. This work investigates a field of action for implementing a model that takes advantage of previously modified MHs by learning how to boost the performance of the tailoring process. Following this reasoning, we propose a hyper-heuristic model based on Artificial Neural Networks (ANNs) trained with processed sequences of heuristics to identify patterns that one can use to generate better MHs. We prove the feasibility of this model by comparing the results against generic MHs and other approaches that tailor unfolded MHs. Our results evidenced that the proposed model outperformed an average of 84 % of all scenarios; in particular, 89 % of basic and 77 % of unfolded approaches. Plus, we highlight the configurable capability of the proposed model, as it shows to be exceptionally versatile in regards to the computational budget, generating good results even with limited resources.
{"title":"A Primary Study on Hyper-Heuristics Powered by Artificial Neural Networks for Customising Population-based Metaheuristics in Continuous Optimisation Problems","authors":"Jose M. Tapia-Avitia, J. M. Cruz-Duarte, I. Amaya, J. C. Ortíz-Bayliss, H. Terashima-Marín, N. Pillay","doi":"10.1109/CEC55065.2022.9870275","DOIUrl":"https://doi.org/10.1109/CEC55065.2022.9870275","url":null,"abstract":"Metaheuristics (MHs) are proven powerful algorithms for solving non-linear optimisation problems over discrete, continuous, or mixed domains. Applications have ranged from basic sciences to applied technologies. Nowadays, the literature contains plenty of MHs based on exceptional ideas, but often, they are just recombining elements from other techniques. An alternative approach is to follow a standard model that customises population-based MHs, utilising simple heuristics extracted from well-known MHs. Different approaches have explored the combination of such simple heuristics, generating excellent results compared to the generic MHs. Nevertheless, they present limitations due to the nature of the metaheuristic used to study the heuristic space. This work investigates a field of action for implementing a model that takes advantage of previously modified MHs by learning how to boost the performance of the tailoring process. Following this reasoning, we propose a hyper-heuristic model based on Artificial Neural Networks (ANNs) trained with processed sequences of heuristics to identify patterns that one can use to generate better MHs. We prove the feasibility of this model by comparing the results against generic MHs and other approaches that tailor unfolded MHs. Our results evidenced that the proposed model outperformed an average of 84 % of all scenarios; in particular, 89 % of basic and 77 % of unfolded approaches. Plus, we highlight the configurable capability of the proposed model, as it shows to be exceptionally versatile in regards to the computational budget, generating good results even with limited resources.","PeriodicalId":153241,"journal":{"name":"2022 IEEE Congress on Evolutionary Computation (CEC)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125359983","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-18DOI: 10.1109/CEC55065.2022.9870322
Atiya Masood, Gang Chen, Yi Mei, Harith Al-Sahaf, Mengjie Zhang
Job Shop Scheduling (JSS) is an important real-world problem. However, the problem is challenging because of many conflicting objectives and the complexity of production flows. Genetic programming-based hyper-heuristic (GP-HH) is a useful approach for automatically evolving effective dispatching rules for many-objective JSS. However, the evolved Pareto-front is highly irregular, seriously affecting the effectiveness of GP-HH. Although the reference points method is one of the most prominent and efficient methods for diversity maintenance in many-objective problems, it usually uses a uniform distribution of reference points which is only appropriate for a regular Pareto-front. In fact, some reference points may never be linked to any Pareto-optimal solutions, rendering them useless. These useless reference points can significantly impact the performance of any reference-point-based many-objective optimization algorithms such as NSGA-III. This paper proposes a new reference point adaption process that explicitly constructs the distribution model using Gaussian process to effectively reduce the number of useless reference points to a low level, enabling a close match between reference points and the distribution of Pareto-optimal solutions. We incorporate this mechanism into NSGA-III to build a new algorithm called MARP-NSGA-III which is compared experimentally to several popular many-objective algorithms. Experiment results on a large collection of many-objective benchmark JSS instances clearly show that MARP-NSGA-III can significantly improve the performance by using our Gaussian Process-based reference point adaptation mechanism.
{"title":"Genetic Programming Hyper-heuristic with Gaussian Process-based Reference Point Adaption for Many-Objective Job Shop Scheduling","authors":"Atiya Masood, Gang Chen, Yi Mei, Harith Al-Sahaf, Mengjie Zhang","doi":"10.1109/CEC55065.2022.9870322","DOIUrl":"https://doi.org/10.1109/CEC55065.2022.9870322","url":null,"abstract":"Job Shop Scheduling (JSS) is an important real-world problem. However, the problem is challenging because of many conflicting objectives and the complexity of production flows. Genetic programming-based hyper-heuristic (GP-HH) is a useful approach for automatically evolving effective dispatching rules for many-objective JSS. However, the evolved Pareto-front is highly irregular, seriously affecting the effectiveness of GP-HH. Although the reference points method is one of the most prominent and efficient methods for diversity maintenance in many-objective problems, it usually uses a uniform distribution of reference points which is only appropriate for a regular Pareto-front. In fact, some reference points may never be linked to any Pareto-optimal solutions, rendering them useless. These useless reference points can significantly impact the performance of any reference-point-based many-objective optimization algorithms such as NSGA-III. This paper proposes a new reference point adaption process that explicitly constructs the distribution model using Gaussian process to effectively reduce the number of useless reference points to a low level, enabling a close match between reference points and the distribution of Pareto-optimal solutions. We incorporate this mechanism into NSGA-III to build a new algorithm called MARP-NSGA-III which is compared experimentally to several popular many-objective algorithms. Experiment results on a large collection of many-objective benchmark JSS instances clearly show that MARP-NSGA-III can significantly improve the performance by using our Gaussian Process-based reference point adaptation mechanism.","PeriodicalId":153241,"journal":{"name":"2022 IEEE Congress on Evolutionary Computation (CEC)","volume":"513 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122214010","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-18DOI: 10.1109/CEC55065.2022.9870349
Naoya Yatsu, Hiroki Shiraishi, Hiroyuki Sato, K. Takadama
This paper focuses on the matching mechanism of Learning Classifier System (LCS) in a continuous space and proposes a novel matching mechanism based on Gaussian distribution. This mechanism can match the “area” instead of the “point (one value)” in the continuous space unlike the conventional LCS such as XCSR (XCS with Continuous-Valued Inputs). Such an area matching contributes to generating the rules (called classifiers) with less-overlapped with other rules. Concretely, the proposed area matching mechanism employed in XCSR using VAE can generate appropriate classifiers for latent variables with high-dimensional inputs by VAE and create a human-interpretable observation space of human-interpretable classifiers. Since the latent variable in VAE is followed by Gaus-sian distribution, the following three matching mechanisms are compared: (i) the (single) point matching that selects the classifier which condition covers the mean of Gaussian distribution M; (ii) the multiple points matching that selects the classifier which condition covers the data sampled from Gaussian distribution (M, u); and (iii) the area matching that selects the classifier which condition roughly covers a certain area of Gaussian distribution (M, o). Through the intensive experiments on the high dimension maze problem, the following implications have been revealed: (1) the point matching in XCSR with VAE generates the ambiguous classifiers which conditions are overlapped with the other classifiers with the different action; (2) the sampling multiple points matching in XCSR with VAE has a potential of generating the less-overlapped classifiers by improving the data set through sampling. (3) the proposed area matching can generate the less-overlapped classifiers with the same learning steps, which corresponds to the time of the point matching.
{"title":"XCSR with VAE using Gaussian Distribution Matching: From Point to Area Matching in Latent Space for Less-overlapped Rule Generation in Observation Space","authors":"Naoya Yatsu, Hiroki Shiraishi, Hiroyuki Sato, K. Takadama","doi":"10.1109/CEC55065.2022.9870349","DOIUrl":"https://doi.org/10.1109/CEC55065.2022.9870349","url":null,"abstract":"This paper focuses on the matching mechanism of Learning Classifier System (LCS) in a continuous space and proposes a novel matching mechanism based on Gaussian distribution. This mechanism can match the “area” instead of the “point (one value)” in the continuous space unlike the conventional LCS such as XCSR (XCS with Continuous-Valued Inputs). Such an area matching contributes to generating the rules (called classifiers) with less-overlapped with other rules. Concretely, the proposed area matching mechanism employed in XCSR using VAE can generate appropriate classifiers for latent variables with high-dimensional inputs by VAE and create a human-interpretable observation space of human-interpretable classifiers. Since the latent variable in VAE is followed by Gaus-sian distribution, the following three matching mechanisms are compared: (i) the (single) point matching that selects the classifier which condition covers the mean of Gaussian distribution M; (ii) the multiple points matching that selects the classifier which condition covers the data sampled from Gaussian distribution (M, u); and (iii) the area matching that selects the classifier which condition roughly covers a certain area of Gaussian distribution (M, o). Through the intensive experiments on the high dimension maze problem, the following implications have been revealed: (1) the point matching in XCSR with VAE generates the ambiguous classifiers which conditions are overlapped with the other classifiers with the different action; (2) the sampling multiple points matching in XCSR with VAE has a potential of generating the less-overlapped classifiers by improving the data set through sampling. (3) the proposed area matching can generate the less-overlapped classifiers with the same learning steps, which corresponds to the time of the point matching.","PeriodicalId":153241,"journal":{"name":"2022 IEEE Congress on Evolutionary Computation (CEC)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125192152","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-18DOI: 10.1109/CEC55065.2022.9870422
Meera Ramadas, A. Abraham
Soil Moisture aid analysts in study of soil science, agriculture and hydrology. Satellite imagery for soil moisture estimation is recorded through earth satellites. By segmenting these satellite imageries based on soil moisture content, we can effortlessly identify regions of wetter condition and regions of dry condition. Differential evolution (DE) is a popular evolutionary approach that is used to optimize problems like image segmentation. In this work, an Advanced Differential Evolution (aDE) technique is introduced which has enhanced performance in comparison to traditional DE approach. This approach is combined with Renyi's entropy for performing multilevel segmentation on the imagery. The resultant segmented images obtained on using the proposed technique is of enhanced quality.
{"title":"Segregating Satellite Imagery Based on Soil Moisture Level Using Advanced Differential Evolutionary Multilevel Segmentation","authors":"Meera Ramadas, A. Abraham","doi":"10.1109/CEC55065.2022.9870422","DOIUrl":"https://doi.org/10.1109/CEC55065.2022.9870422","url":null,"abstract":"Soil Moisture aid analysts in study of soil science, agriculture and hydrology. Satellite imagery for soil moisture estimation is recorded through earth satellites. By segmenting these satellite imageries based on soil moisture content, we can effortlessly identify regions of wetter condition and regions of dry condition. Differential evolution (DE) is a popular evolutionary approach that is used to optimize problems like image segmentation. In this work, an Advanced Differential Evolution (aDE) technique is introduced which has enhanced performance in comparison to traditional DE approach. This approach is combined with Renyi's entropy for performing multilevel segmentation on the imagery. The resultant segmented images obtained on using the proposed technique is of enhanced quality.","PeriodicalId":153241,"journal":{"name":"2022 IEEE Congress on Evolutionary Computation (CEC)","volume":"109 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122493697","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-18DOI: 10.1109/CEC55065.2022.9870325
Eduardo Lacerda, F. Lezama, J. Soares, B. Canizes, Z. Vale
This article evaluates the performance of different metaheuristics (evolutionary algorithms) solving a cost mini-mization problem in demand response contract markets. The problem considers a contract market in which a distribution system operator (DSO) requests flexibility from aggregators with DR capabilities. We include a network validation approach in the evaluation of solutions, i.e., the DSO determines losses and voltage limit violations depending on the location of aggregators in the network. The validation of the network increases the complexity of the objective function since new network constraints are included in the formulation. Therefore, we advocate the use of metaheuristic optimization and a simulation procedure to overcome this issue. We compare different evolutionary algorithms, including the well-known differential evolution and other two more recent algorithms, the vortex search and the hybrid-adaptive differential evolution with decay function. Results demonstrate the effectiveness of these approaches in solving the proposed complex model under a realistic case study.
{"title":"Metaheuristic Optimization Solving Demand Response Contract Markets with Network Validation","authors":"Eduardo Lacerda, F. Lezama, J. Soares, B. Canizes, Z. Vale","doi":"10.1109/CEC55065.2022.9870325","DOIUrl":"https://doi.org/10.1109/CEC55065.2022.9870325","url":null,"abstract":"This article evaluates the performance of different metaheuristics (evolutionary algorithms) solving a cost mini-mization problem in demand response contract markets. The problem considers a contract market in which a distribution system operator (DSO) requests flexibility from aggregators with DR capabilities. We include a network validation approach in the evaluation of solutions, i.e., the DSO determines losses and voltage limit violations depending on the location of aggregators in the network. The validation of the network increases the complexity of the objective function since new network constraints are included in the formulation. Therefore, we advocate the use of metaheuristic optimization and a simulation procedure to overcome this issue. We compare different evolutionary algorithms, including the well-known differential evolution and other two more recent algorithms, the vortex search and the hybrid-adaptive differential evolution with decay function. Results demonstrate the effectiveness of these approaches in solving the proposed complex model under a realistic case study.","PeriodicalId":153241,"journal":{"name":"2022 IEEE Congress on Evolutionary Computation (CEC)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131623152","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-18DOI: 10.1109/CEC55065.2022.9870407
Ray Lim, Abhishek Gupta, Y. Ong
Transfer evolutionary optimization (TrEO) has emerged as a computational paradigm to leverage related problem-solving information from various source tasks to boost convergence rates in a target task. State-of-the-art Tr EO algorithms have utilized a source-target similarity capture method with probabilistic priors that grants the ability to reduce negative transfers. A recent work makes use of an additional solution representation learning module to induce high ordinal correlation between source and target objective functions through source-to-target search space mappings, with the aim of promoting positive transfers between them. However, current implementations of this approach are found to be data-intensive - calling for all generated source data to be cached - leading to high storage costs in practice. As an alternative, this paper investigates the feasibility of a data-lean variant of the aforesaid approach, labeled as (1, G)-TrEO, in which only the first and final (Gth) generations of source data are used for solution representation learning and transfer. We conduct experimental analyses of (1, G)-TrEO using multi-objective benchmark functions as well as a practical example in vehicle crashworthiness design. Our results show that a simple data-lean transfer optimizer is able to achieve competitive performance. While this paper presents a first investigation of (1, G)-TrEO, we hope that the findings would inspire future forms of data-lean TrEO algorithms.
{"title":"An Initial Investigation of Data-Lean Transfer Evolutionary Optimization with Probabilistic Priors","authors":"Ray Lim, Abhishek Gupta, Y. Ong","doi":"10.1109/CEC55065.2022.9870407","DOIUrl":"https://doi.org/10.1109/CEC55065.2022.9870407","url":null,"abstract":"Transfer evolutionary optimization (TrEO) has emerged as a computational paradigm to leverage related problem-solving information from various source tasks to boost convergence rates in a target task. State-of-the-art Tr EO algorithms have utilized a source-target similarity capture method with probabilistic priors that grants the ability to reduce negative transfers. A recent work makes use of an additional solution representation learning module to induce high ordinal correlation between source and target objective functions through source-to-target search space mappings, with the aim of promoting positive transfers between them. However, current implementations of this approach are found to be data-intensive - calling for all generated source data to be cached - leading to high storage costs in practice. As an alternative, this paper investigates the feasibility of a data-lean variant of the aforesaid approach, labeled as (1, G)-TrEO, in which only the first and final (Gth) generations of source data are used for solution representation learning and transfer. We conduct experimental analyses of (1, G)-TrEO using multi-objective benchmark functions as well as a practical example in vehicle crashworthiness design. Our results show that a simple data-lean transfer optimizer is able to achieve competitive performance. While this paper presents a first investigation of (1, G)-TrEO, we hope that the findings would inspire future forms of data-lean TrEO algorithms.","PeriodicalId":153241,"journal":{"name":"2022 IEEE Congress on Evolutionary Computation (CEC)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128039609","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-18DOI: 10.1109/CEC55065.2022.9870287
Yoshihiko Ozaki, Shintaro Takenaga, Masaki Onishi
Hyperparameter optimization (HPO) is a compu-tationally expensive blackbox optimization problem to maximize the performance of a machine learning model by tuning the model hyperparameters. Conventionally, global search has been widely adopted rather than local search to address HPO. In this study, we investigate whether this conventional choice is reasonable by empirically comparing popular global and local search methods as applied to HPO problems. The numerical results demonstrate that local search methods consistently achieve results that are comparable to or better than those of the global search methods, i.e., local search is a more reasonable choice for HPO. We also report the findings of detailed analyses of the experimental data conducted to understand how each method functions and the objective landscapes of HPO.
{"title":"Global Search versus Local Search in Hyperparameter Optimization","authors":"Yoshihiko Ozaki, Shintaro Takenaga, Masaki Onishi","doi":"10.1109/CEC55065.2022.9870287","DOIUrl":"https://doi.org/10.1109/CEC55065.2022.9870287","url":null,"abstract":"Hyperparameter optimization (HPO) is a compu-tationally expensive blackbox optimization problem to maximize the performance of a machine learning model by tuning the model hyperparameters. Conventionally, global search has been widely adopted rather than local search to address HPO. In this study, we investigate whether this conventional choice is reasonable by empirically comparing popular global and local search methods as applied to HPO problems. The numerical results demonstrate that local search methods consistently achieve results that are comparable to or better than those of the global search methods, i.e., local search is a more reasonable choice for HPO. We also report the findings of detailed analyses of the experimental data conducted to understand how each method functions and the objective landscapes of HPO.","PeriodicalId":153241,"journal":{"name":"2022 IEEE Congress on Evolutionary Computation (CEC)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133412060","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：