Guy Shani - Academia.edu (original) (raw)
Papers by Guy Shani
Proceedings of the International Symposium on Combinatorial Search, Jul 17, 2022
Proceedings of the International Conference on Automated Planning and Scheduling
Collaborative privacy-preserving planning (CPPP) is a multi-agent planning task in which agents n... more Collaborative privacy-preserving planning (CPPP) is a multi-agent planning task in which agents need to achieve a common set of goals without revealing certain private information. In many CPPP algorithms the individual agents reason about a projection of the multiagent problem onto a single-agent classical planning problem. For example, an agent can plan as if it controls the public actions of other agents, ignoring their unknown private preconditions and effects, and use the cost of this plan as a heuristic for the cost of the full, multi-agent plan. Using such a projection, however, ignores some dependencies between agents’ public actions. In particular, it does not contain dependencies between actions of other agents caused by their private facts. We propose a projection in which these private dependencies are maintained. The benefit of our dependency-preserving projection is demonstrated by using it to produce high level plans in a new privacy preserving planner that is able to...
Proceedings of the International Conference on Automated Planning and Scheduling
Collaborative Multi-Agent Planning (MAP) under uncertainty with partial observability is a notori... more Collaborative Multi-Agent Planning (MAP) under uncertainty with partial observability is a notoriously difficult problem. Such MAP problems are often modeled as DecPOMDPs, or its qualitative variant, QDec-POMDP, which is essentially a MAP version of contingent planning. The QDecPOMDP model was introduced with the hope that its simpler, non-probabilistic structure will allow for better scalability. Indeed, at least with deterministic actions, the recent IMAP algorithm scales much better than comparable DecPOMDP algorithms (Bazinin and Shani 2018). In this work we suggest a new approach to solving Deterministic QDecPOMDPs based on problem factoring. First, we find a solution to a MAP problem where the results of any observation is available to all agents. This is essentially a single-agent planning problem for the entire team. Then, we project the solution tree into sub-trees, one per agent, and let each agent transform its projected tree into a legal local tree. If all agents succeed...
Proceedings of the AAAI Conference on Artificial Intelligence
In contingent planning under partial observability with sensing actions, agents actively use sens... more In contingent planning under partial observability with sensing actions, agents actively use sensing to discover meaningful facts about the world. For this class of problems the solution can be represented as a plan tree, branching on various possible observations. Recent successful approaches translate the partially observable contingent problem into a non-deterministic fully observable problem, and then use a planner for non-deterministic planning. While this approach has been successful in many domains, the translation may become very large, encumbering the task of the non-deterministic planner. In this paper we suggest a different approach - using an online contingent solver repeatedly to construct a plan tree. We execute the plan returned by the online solver until the next observation action, and then branch on the possible observed values, and replan for every branch independently. In many cases a plan tree can be exponential in the number of state variables, but still, the t...
Abstract—Recently, POMDP solvers have shown the ability to scale up significantly using domain st... more Abstract—Recently, POMDP solvers have shown the ability to scale up significantly using domain structure, such as factored representations. In many domains the agent is required to complete a set of independent tasks. We propose to decompose a factored POMDP into a set of restricted POMDPs over subsets of task relevant state variables. We solve each such model independently, acquiring a value function. The combination of the value functions of the restricted POMDPs is then used to form a policy for the complete POMDP. We explain the process of identifying variables that correspond to tasks, and how to create a model restricted to a single task, or to a subset of tasks. We demonstrate our approach on a number of benchmarks from the factored POMDP literature, showing that our methods are applicable to models with more than 100 state variables.
Introduction—The inability to recover from unexpected lateral loss of balance may be particularly... more Introduction—The inability to recover from unexpected lateral loss of balance may be particularly relevant to the problem of falls.Aim—We aimed to explore whether different kinematic patterns and strategies occur in the first recovery step in single-step trials when single step was required to recover from fall and in the multiple-step trials, when more than one step were required to recover from fall. In addition, in the multiple-step trials we examined kinematic patterns of balance recovery where extra steps were needed to recover balance. Methods—Eighty-four older adults (79.3±5.2 years) were exposed to announced right/left perturbations in standing that were gradually increased to trigger a recovery stepping response. We performed kinematic analysis of the first recovery step of all single-step and multiple-step trials for each participant and of total balance recovery in the multiple-step trial.Results—Kinematic patterns and strategies of the first recovery step in the single-s...
Alternative RNA splicing results in multiple transcripts of the same gene, possibly encoding for ... more Alternative RNA splicing results in multiple transcripts of the same gene, possibly encoding for different protein isoforms with different protein domains and functionalities. Whereas it is possible to manually determine the effect of a specific alternative splicing event on the domain composition of a particular encoded protein, the process requires the tedious integration of several data sources; it is therefore error prone and its implementation is not feasible for genome-wide characterization of domains affected by differential splicing. To fulfill the need for an automated solution, we developed the Domain Change Presenter (DoChaP), a web server for the visualization of the exon–domain association. DoChaP visualizes all transcripts of a given gene, the domains of the proteins that they encode, and the exons encoding each domain. The visualization enables a comparison between the transcripts and between the protein isoforms they encode for. The organization and visual presentati...
BMC Geriatrics, 2021
Background Balance control, and specifically balance reactive responses that contribute to mainta... more Background Balance control, and specifically balance reactive responses that contribute to maintaining balance when balance is lost unexpectedly, is impaired in older people. This leads to an increased fall risk and injurious falls. Improving balance reactive responses is one of the goals in fall-prevention training programs. Perturbation training during standing or treadmill walking that specifically challenges the balance reactive responses has shown very promising results; however, only older people who are able to perform treadmill walking can participate in these training regimes. Thus, we aimed to develop, build, and pilot a mechatronic Perturbation Stationary Bicycle Robotic system (i.e., PerStBiRo) that can challenge balance while sitting on a stationary bicycle, with the aim of improving balance proactive and reactive control. Methods This paper describes the development, and building of the PerStBiRo using stationary bicycles. In addition, we conducted a pilot randomized c...
BMC Geriatrics, 2020
Background Step-recovery responses are critical in preventing falls when balance is lost unexpect... more Background Step-recovery responses are critical in preventing falls when balance is lost unexpectedly. We investigated the kinematics and strategies of balance recovery in older adults with a varying history of falls. Methods In a laboratory study, 51 non-fallers (NFs), 20 one-time fallers (OFs), and 12 recurrent-fallers (RFs) were exposed to random right/left unannounced underfoot perturbations in standing of increasing magnitude. The stepping strategies and kinematics across an increasing magnitude of perturbations and the single- and multiple-step threshold trials, i.e., the lowest perturbation magnitude to evoke single step and multiple steps, respectively, were analyzed. Fall efficacy (FES) and self-reported lower-extremity function were also assessed. Results OFs had significantly lower single- and multiple-step threshold levels than NFs; the recovery-step kinematics were similar. Surprisingly, RFs did not differ from NFs in either threshold. The kinematics in the single-step ...
EPiC Series in Computing
Software vulnerabilities in organizational computer networks can be leveraged by an attacker to g... more Software vulnerabilities in organizational computer networks can be leveraged by an attacker to gain access to sensitive information. As fixing all vulnerabilities requires much effort, it is critical to rank the possible fixes by their importance. Centrality measures over logical attack graphs, or over the network connectivity graph, often provide a scalable method for finding the most critical vulnerabilities.In this paper we suggest an analysis of the planning graph, originating in classical planning, as an alternative for the logical attack graph, to improve the ranking produced by centrality measures. The planning graph also allows us to enumerate the set of possible attack plans, and hence, directly count the number of attacks that use a given vulnerability. We evaluate a set of centrality-based ranking measures over the logical attack graph and the planning graph, showing that metrics computed over the planning graph reduce more rapidly the set of shortest attack plans.
Gerontology, 2020
Introduction: Many falls in older adults occur during walking and result in lateral falls. The ab... more Introduction: Many falls in older adults occur during walking and result in lateral falls. The ability to perform a recovery step after balance perturbation determines whether a fall will occur. Aim: To investigate age-related changes in first recovery step kinematics and kinematic adaptations over a wide range of lateral perturbation magnitudes while walking. Methods: Thirty-five old (78.5 ± 5 years) and 19 young adults (26.0 ± 0.8 years) walked at their preferred walking speed on a treadmill. While walking, the subjects were exposed to announced right/left perturbations in different phases of the gait cycle that were gradually increased in order to trigger a recovery stepping response. The subjects were instructed to react naturally and try to avoid falling. Kinematic analysis was performed to analyze the first recovery step parameters (e.g., step initiation, swing duration, step length, and the estimated distance of the center of mass from the base of support [dBoS]). Results: Co...
Neurorehabilitation and Neural Repair, 2019
Background: Reactive balance responses are critical for fall prevention. Perturbation-based balan... more Background: Reactive balance responses are critical for fall prevention. Perturbation-based balance training (PBBT) has shown a positive effect in reducing the risk of falls among older adults and persons with Parkinson’s disease. Objective: To explore the effect of a short-term PBBT on reactive balance responses, performance-based measures of balance and gait and balance confidence. Methods: Thirty-four moderate-high functioning, subacute persons with stroke (PwS) (lower extremity Fugl-Meyer score 29.2 ± 4.3; Berg Balance Scale [BBS] score 43.8 ± 9.5, 42.0 ± 18.7 days after stroke onset) hospitalized in a rehabilitation setting were randomly allocated to PBBT (n = 18) and weight shifting and gait training (WS>) (n = 16). Both groups received 12 training sessions, 30 minutes each, for a period of 2.5 weeks. PBBT included unexpected balance perturbations during standing and treadmill walking, WS> included weight shifting in standing and treadmill walking without perturbations. Th...
Phytopathology®, 2018
Many plant diseases have distinct visual symptoms, which can be used to identify and classify the... more Many plant diseases have distinct visual symptoms, which can be used to identify and classify them correctly. This article presents a potato disease classification algorithm that leverages these distinct appearances and advances in computer vision made possible by deep learning. The algorithm uses a deep convolutional neural network, training it to classify the tubers into five classes: namely, four disease classes and a healthy potato class. The database of images used in this study, containing potato tubers of different cultivars, sizes, and diseases, was acquired, classified, and labeled manually by experts. The models were trained over different train-test splits to better understand the amount of image data needed to apply deep learning for such classification tasks. The models were tested over a data set of images taken using standard low-cost RGB (red, green, and blue) sensors and were tagged by experts, demonstrating high classification accuracy. This is the first article to...
Phenotyping is the task of measuring plant attributes for analyzing the current state of the plan... more Phenotyping is the task of measuring plant attributes for analyzing the current state of the plant. In agriculture, phenotyping can be used to make decisions concerning the management of crops, such as the watering policy, or whether to spray for a certain pest. Currently, large scale phenotyping in fields is typically done using manual labor, which is a costly, low throughput process. Researchers often advocate the use of automated systems for phenotyping, relying on the use of sensors for making measurements. The recent rise of low cost, yet reasonably accurate, RGB-D sensors has opened the way for using these sensors in field phenotyping applications. In this paper, we investigate the applicability of 4 different RGB-D sensors for this task. We conduct an outdoor experiment, measuring plant attribute in various distances and light conditions. Our results show that modern RGB-D sensors, in particular, the Intel D435 sensor, provides a viable tool for close range phenotyping tasks ...
ACM Transactions on Intelligent Systems and Technology, 2016
Recommender systems (RS) can now be found in many commercial Web sites, often presenting customer... more Recommender systems (RS) can now be found in many commercial Web sites, often presenting customers with a short list of additional products that they might purchase. Many commercial sites do not typically have the ability and resources to develop their own system and may outsource the RS to a third party. This had led to the growth of a recommendation as a service industry, where companies, referred to as RS providers, provide recommendation services. These companies must carefully balance the cost of building recommendation models and the payment received from the e-business, as these payments are expected to be low. In such a setting, restricting the computational time required for model building is critical for the RS provider to be profitable. In this article, we propose anytime algorithms as an attractive method for balancing computational time and the recommendation model performance, thus tackling the RS provider problem. In an anytime setting, an algorithm can be stopped aft...
Proceedings of the International Conference on Agents and Artificial Intelligence, 2015
Item-based Collaborative Filtering (CF) models offer good recommendations with low latency. Still... more Item-based Collaborative Filtering (CF) models offer good recommendations with low latency. Still, constructing such models is often slow, requiring the comparison of all item pairs, and then caching for each item the list of most similar items. In this paper we suggest methods for reducing the number of item pairs comparisons, through simple clustering, where similar items tend to be in the same cluster. We propose two methods, one that uses Locality Sensitive Hashing (LSH), and another that uses the item consumption cardinality. We evaluate the two methods demonstrating the cardinality based method reduce the computation time dramatically without damage the accuracy.
Proceedings of the International Conference on Automated Planning and Scheduling
Collaborative privacy preserving planning (cppp) has gained much attention in the past decade. To... more Collaborative privacy preserving planning (cppp) has gained much attention in the past decade. To date, cppp has focused on domains with deterministic action effects. In this paper, we extend cppp to domains with stochastic action effects. We show how such environments can be modeled as an mdp. We then focus on the popular Real-Time Dynamic Programming (RTDP) algorithm for computing value functions for mdps, extending it to the stochastic cppp setting. We provide two versions of RTDP: a complete version identical to executing centralized RTDP, and an approximate version that sends significantly fewer messages and computes competitive policies in practice. We experiment on domains adapted from the deterministic cppp literature.
Vietnam Journal of Computer Science, 2019
An escape room is a physical puzzle solving game, where participants solve a series of riddles wi... more An escape room is a physical puzzle solving game, where participants solve a series of riddles within a limited time to exit a locked room. Escape rooms differ in their theme, environment, and difficulty, and people hence often differ on their preferences over escape rooms. As such, recommendation systems can help people in deciding which room to visit. In this paper, we describe the properties of the escape rooms recommendation problem, with respect to other popular recommendation problems. We describe a dataset of reviews collected within a current system. We provide an empirical comparison between a set of recommendation algorithms over two problems, top-N recommendation and rating prediction. In both cases, a KNN method performed the best.
Proceedings of the International Conference on Automated Planning and Scheduling
In contingent planning problems, agents have partial information about their state anduse sensing... more In contingent planning problems, agents have partial information about their state anduse sensing actions to learn the value of some variables.When sensing and actuation are separated, plans for such problems can often be viewed as a tree of sensing actions, separated by conformant plans consisting of non-sensing actions that enable the execution of the next sensing action. This leads us to propose a heuristic, online method for contingent planning which focuses on identifying thenext useful sensing action. The key part of our planner is a novel landmarks-based heuristic for selecting the next sensing action, together with a projection method that uses classical planning to solve the intermediate conformant planning problems.This allows our planner to operate without an explicit model of belief space or the use of existing translation techniques,both of which can require exponential space. The resulting Heuristic Contingent Planner (HCP) solves many more problems than state-of-the-a...
Proceedings of the International Symposium on Combinatorial Search, Jul 17, 2022
Proceedings of the International Conference on Automated Planning and Scheduling
Collaborative privacy-preserving planning (CPPP) is a multi-agent planning task in which agents n... more Collaborative privacy-preserving planning (CPPP) is a multi-agent planning task in which agents need to achieve a common set of goals without revealing certain private information. In many CPPP algorithms the individual agents reason about a projection of the multiagent problem onto a single-agent classical planning problem. For example, an agent can plan as if it controls the public actions of other agents, ignoring their unknown private preconditions and effects, and use the cost of this plan as a heuristic for the cost of the full, multi-agent plan. Using such a projection, however, ignores some dependencies between agents’ public actions. In particular, it does not contain dependencies between actions of other agents caused by their private facts. We propose a projection in which these private dependencies are maintained. The benefit of our dependency-preserving projection is demonstrated by using it to produce high level plans in a new privacy preserving planner that is able to...
Proceedings of the International Conference on Automated Planning and Scheduling
Collaborative Multi-Agent Planning (MAP) under uncertainty with partial observability is a notori... more Collaborative Multi-Agent Planning (MAP) under uncertainty with partial observability is a notoriously difficult problem. Such MAP problems are often modeled as DecPOMDPs, or its qualitative variant, QDec-POMDP, which is essentially a MAP version of contingent planning. The QDecPOMDP model was introduced with the hope that its simpler, non-probabilistic structure will allow for better scalability. Indeed, at least with deterministic actions, the recent IMAP algorithm scales much better than comparable DecPOMDP algorithms (Bazinin and Shani 2018). In this work we suggest a new approach to solving Deterministic QDecPOMDPs based on problem factoring. First, we find a solution to a MAP problem where the results of any observation is available to all agents. This is essentially a single-agent planning problem for the entire team. Then, we project the solution tree into sub-trees, one per agent, and let each agent transform its projected tree into a legal local tree. If all agents succeed...
Proceedings of the AAAI Conference on Artificial Intelligence
In contingent planning under partial observability with sensing actions, agents actively use sens... more In contingent planning under partial observability with sensing actions, agents actively use sensing to discover meaningful facts about the world. For this class of problems the solution can be represented as a plan tree, branching on various possible observations. Recent successful approaches translate the partially observable contingent problem into a non-deterministic fully observable problem, and then use a planner for non-deterministic planning. While this approach has been successful in many domains, the translation may become very large, encumbering the task of the non-deterministic planner. In this paper we suggest a different approach - using an online contingent solver repeatedly to construct a plan tree. We execute the plan returned by the online solver until the next observation action, and then branch on the possible observed values, and replan for every branch independently. In many cases a plan tree can be exponential in the number of state variables, but still, the t...
Abstract—Recently, POMDP solvers have shown the ability to scale up significantly using domain st... more Abstract—Recently, POMDP solvers have shown the ability to scale up significantly using domain structure, such as factored representations. In many domains the agent is required to complete a set of independent tasks. We propose to decompose a factored POMDP into a set of restricted POMDPs over subsets of task relevant state variables. We solve each such model independently, acquiring a value function. The combination of the value functions of the restricted POMDPs is then used to form a policy for the complete POMDP. We explain the process of identifying variables that correspond to tasks, and how to create a model restricted to a single task, or to a subset of tasks. We demonstrate our approach on a number of benchmarks from the factored POMDP literature, showing that our methods are applicable to models with more than 100 state variables.
Introduction—The inability to recover from unexpected lateral loss of balance may be particularly... more Introduction—The inability to recover from unexpected lateral loss of balance may be particularly relevant to the problem of falls.Aim—We aimed to explore whether different kinematic patterns and strategies occur in the first recovery step in single-step trials when single step was required to recover from fall and in the multiple-step trials, when more than one step were required to recover from fall. In addition, in the multiple-step trials we examined kinematic patterns of balance recovery where extra steps were needed to recover balance. Methods—Eighty-four older adults (79.3±5.2 years) were exposed to announced right/left perturbations in standing that were gradually increased to trigger a recovery stepping response. We performed kinematic analysis of the first recovery step of all single-step and multiple-step trials for each participant and of total balance recovery in the multiple-step trial.Results—Kinematic patterns and strategies of the first recovery step in the single-s...
Alternative RNA splicing results in multiple transcripts of the same gene, possibly encoding for ... more Alternative RNA splicing results in multiple transcripts of the same gene, possibly encoding for different protein isoforms with different protein domains and functionalities. Whereas it is possible to manually determine the effect of a specific alternative splicing event on the domain composition of a particular encoded protein, the process requires the tedious integration of several data sources; it is therefore error prone and its implementation is not feasible for genome-wide characterization of domains affected by differential splicing. To fulfill the need for an automated solution, we developed the Domain Change Presenter (DoChaP), a web server for the visualization of the exon–domain association. DoChaP visualizes all transcripts of a given gene, the domains of the proteins that they encode, and the exons encoding each domain. The visualization enables a comparison between the transcripts and between the protein isoforms they encode for. The organization and visual presentati...
BMC Geriatrics, 2021
Background Balance control, and specifically balance reactive responses that contribute to mainta... more Background Balance control, and specifically balance reactive responses that contribute to maintaining balance when balance is lost unexpectedly, is impaired in older people. This leads to an increased fall risk and injurious falls. Improving balance reactive responses is one of the goals in fall-prevention training programs. Perturbation training during standing or treadmill walking that specifically challenges the balance reactive responses has shown very promising results; however, only older people who are able to perform treadmill walking can participate in these training regimes. Thus, we aimed to develop, build, and pilot a mechatronic Perturbation Stationary Bicycle Robotic system (i.e., PerStBiRo) that can challenge balance while sitting on a stationary bicycle, with the aim of improving balance proactive and reactive control. Methods This paper describes the development, and building of the PerStBiRo using stationary bicycles. In addition, we conducted a pilot randomized c...
BMC Geriatrics, 2020
Background Step-recovery responses are critical in preventing falls when balance is lost unexpect... more Background Step-recovery responses are critical in preventing falls when balance is lost unexpectedly. We investigated the kinematics and strategies of balance recovery in older adults with a varying history of falls. Methods In a laboratory study, 51 non-fallers (NFs), 20 one-time fallers (OFs), and 12 recurrent-fallers (RFs) were exposed to random right/left unannounced underfoot perturbations in standing of increasing magnitude. The stepping strategies and kinematics across an increasing magnitude of perturbations and the single- and multiple-step threshold trials, i.e., the lowest perturbation magnitude to evoke single step and multiple steps, respectively, were analyzed. Fall efficacy (FES) and self-reported lower-extremity function were also assessed. Results OFs had significantly lower single- and multiple-step threshold levels than NFs; the recovery-step kinematics were similar. Surprisingly, RFs did not differ from NFs in either threshold. The kinematics in the single-step ...
EPiC Series in Computing
Software vulnerabilities in organizational computer networks can be leveraged by an attacker to g... more Software vulnerabilities in organizational computer networks can be leveraged by an attacker to gain access to sensitive information. As fixing all vulnerabilities requires much effort, it is critical to rank the possible fixes by their importance. Centrality measures over logical attack graphs, or over the network connectivity graph, often provide a scalable method for finding the most critical vulnerabilities.In this paper we suggest an analysis of the planning graph, originating in classical planning, as an alternative for the logical attack graph, to improve the ranking produced by centrality measures. The planning graph also allows us to enumerate the set of possible attack plans, and hence, directly count the number of attacks that use a given vulnerability. We evaluate a set of centrality-based ranking measures over the logical attack graph and the planning graph, showing that metrics computed over the planning graph reduce more rapidly the set of shortest attack plans.
Gerontology, 2020
Introduction: Many falls in older adults occur during walking and result in lateral falls. The ab... more Introduction: Many falls in older adults occur during walking and result in lateral falls. The ability to perform a recovery step after balance perturbation determines whether a fall will occur. Aim: To investigate age-related changes in first recovery step kinematics and kinematic adaptations over a wide range of lateral perturbation magnitudes while walking. Methods: Thirty-five old (78.5 ± 5 years) and 19 young adults (26.0 ± 0.8 years) walked at their preferred walking speed on a treadmill. While walking, the subjects were exposed to announced right/left perturbations in different phases of the gait cycle that were gradually increased in order to trigger a recovery stepping response. The subjects were instructed to react naturally and try to avoid falling. Kinematic analysis was performed to analyze the first recovery step parameters (e.g., step initiation, swing duration, step length, and the estimated distance of the center of mass from the base of support [dBoS]). Results: Co...
Neurorehabilitation and Neural Repair, 2019
Background: Reactive balance responses are critical for fall prevention. Perturbation-based balan... more Background: Reactive balance responses are critical for fall prevention. Perturbation-based balance training (PBBT) has shown a positive effect in reducing the risk of falls among older adults and persons with Parkinson’s disease. Objective: To explore the effect of a short-term PBBT on reactive balance responses, performance-based measures of balance and gait and balance confidence. Methods: Thirty-four moderate-high functioning, subacute persons with stroke (PwS) (lower extremity Fugl-Meyer score 29.2 ± 4.3; Berg Balance Scale [BBS] score 43.8 ± 9.5, 42.0 ± 18.7 days after stroke onset) hospitalized in a rehabilitation setting were randomly allocated to PBBT (n = 18) and weight shifting and gait training (WS>) (n = 16). Both groups received 12 training sessions, 30 minutes each, for a period of 2.5 weeks. PBBT included unexpected balance perturbations during standing and treadmill walking, WS> included weight shifting in standing and treadmill walking without perturbations. Th...
Phytopathology®, 2018
Many plant diseases have distinct visual symptoms, which can be used to identify and classify the... more Many plant diseases have distinct visual symptoms, which can be used to identify and classify them correctly. This article presents a potato disease classification algorithm that leverages these distinct appearances and advances in computer vision made possible by deep learning. The algorithm uses a deep convolutional neural network, training it to classify the tubers into five classes: namely, four disease classes and a healthy potato class. The database of images used in this study, containing potato tubers of different cultivars, sizes, and diseases, was acquired, classified, and labeled manually by experts. The models were trained over different train-test splits to better understand the amount of image data needed to apply deep learning for such classification tasks. The models were tested over a data set of images taken using standard low-cost RGB (red, green, and blue) sensors and were tagged by experts, demonstrating high classification accuracy. This is the first article to...
Phenotyping is the task of measuring plant attributes for analyzing the current state of the plan... more Phenotyping is the task of measuring plant attributes for analyzing the current state of the plant. In agriculture, phenotyping can be used to make decisions concerning the management of crops, such as the watering policy, or whether to spray for a certain pest. Currently, large scale phenotyping in fields is typically done using manual labor, which is a costly, low throughput process. Researchers often advocate the use of automated systems for phenotyping, relying on the use of sensors for making measurements. The recent rise of low cost, yet reasonably accurate, RGB-D sensors has opened the way for using these sensors in field phenotyping applications. In this paper, we investigate the applicability of 4 different RGB-D sensors for this task. We conduct an outdoor experiment, measuring plant attribute in various distances and light conditions. Our results show that modern RGB-D sensors, in particular, the Intel D435 sensor, provides a viable tool for close range phenotyping tasks ...
ACM Transactions on Intelligent Systems and Technology, 2016
Recommender systems (RS) can now be found in many commercial Web sites, often presenting customer... more Recommender systems (RS) can now be found in many commercial Web sites, often presenting customers with a short list of additional products that they might purchase. Many commercial sites do not typically have the ability and resources to develop their own system and may outsource the RS to a third party. This had led to the growth of a recommendation as a service industry, where companies, referred to as RS providers, provide recommendation services. These companies must carefully balance the cost of building recommendation models and the payment received from the e-business, as these payments are expected to be low. In such a setting, restricting the computational time required for model building is critical for the RS provider to be profitable. In this article, we propose anytime algorithms as an attractive method for balancing computational time and the recommendation model performance, thus tackling the RS provider problem. In an anytime setting, an algorithm can be stopped aft...
Proceedings of the International Conference on Agents and Artificial Intelligence, 2015
Item-based Collaborative Filtering (CF) models offer good recommendations with low latency. Still... more Item-based Collaborative Filtering (CF) models offer good recommendations with low latency. Still, constructing such models is often slow, requiring the comparison of all item pairs, and then caching for each item the list of most similar items. In this paper we suggest methods for reducing the number of item pairs comparisons, through simple clustering, where similar items tend to be in the same cluster. We propose two methods, one that uses Locality Sensitive Hashing (LSH), and another that uses the item consumption cardinality. We evaluate the two methods demonstrating the cardinality based method reduce the computation time dramatically without damage the accuracy.
Proceedings of the International Conference on Automated Planning and Scheduling
Collaborative privacy preserving planning (cppp) has gained much attention in the past decade. To... more Collaborative privacy preserving planning (cppp) has gained much attention in the past decade. To date, cppp has focused on domains with deterministic action effects. In this paper, we extend cppp to domains with stochastic action effects. We show how such environments can be modeled as an mdp. We then focus on the popular Real-Time Dynamic Programming (RTDP) algorithm for computing value functions for mdps, extending it to the stochastic cppp setting. We provide two versions of RTDP: a complete version identical to executing centralized RTDP, and an approximate version that sends significantly fewer messages and computes competitive policies in practice. We experiment on domains adapted from the deterministic cppp literature.
Vietnam Journal of Computer Science, 2019
An escape room is a physical puzzle solving game, where participants solve a series of riddles wi... more An escape room is a physical puzzle solving game, where participants solve a series of riddles within a limited time to exit a locked room. Escape rooms differ in their theme, environment, and difficulty, and people hence often differ on their preferences over escape rooms. As such, recommendation systems can help people in deciding which room to visit. In this paper, we describe the properties of the escape rooms recommendation problem, with respect to other popular recommendation problems. We describe a dataset of reviews collected within a current system. We provide an empirical comparison between a set of recommendation algorithms over two problems, top-N recommendation and rating prediction. In both cases, a KNN method performed the best.
Proceedings of the International Conference on Automated Planning and Scheduling
In contingent planning problems, agents have partial information about their state anduse sensing... more In contingent planning problems, agents have partial information about their state anduse sensing actions to learn the value of some variables.When sensing and actuation are separated, plans for such problems can often be viewed as a tree of sensing actions, separated by conformant plans consisting of non-sensing actions that enable the execution of the next sensing action. This leads us to propose a heuristic, online method for contingent planning which focuses on identifying thenext useful sensing action. The key part of our planner is a novel landmarks-based heuristic for selecting the next sensing action, together with a projection method that uses classical planning to solve the intermediate conformant planning problems.This allows our planner to operate without an explicit model of belief space or the use of existing translation techniques,both of which can require exponential space. The resulting Heuristic Contingent Planner (HCP) solves many more problems than state-of-the-a...