580 California St., Suite 400
San Francisco, CA, 94104
Reinforcement learning can greatly benefit from the use of options as a way of encoding recurring behaviours and to foster exploration. An important open problem is how can an agent autonomously learn useful options when solving... more
This paper presents the framework we have developed to classify patients according to the type of hepatitis. To detect the type of virus, once the data have been prepared and encoded in a suitable way, we have extracted the sequential... more
International audienceThis paper presents the framework we have developed to classify patients according to the type of hepatitis. To detect the type of virus, once the data have been prepared and encoded in a suitable way, we have... more
HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or... more
Abstract Finding interesting patterns in time-oriented data can be useful in various domains. In time-oriented clinical data it helps in determining a diagnosis, prescribing therapy and browsing electronic patient records for management... more
In this paper we present a method for the analysis and subsequent clustering of data coming from home monitoring of diabetic patients. Our method aims at characterising the patient behaviour over time in order to be able to cluster... more
In this paper, we describe a novel user interface for the visual definition of temporal abstractions based on a set of intuitive metaphors, which represent both temporal features and logical relations of abstractions.
We study the abstract interpretation of temporal calculi and logics in a general syntax, semantics and abstraction independent setting. This is applied to the z-calculus, a generalization of the ~-calculus with new reversal and... more
This paper considers temporal constraints that can impose a minimum and maximum time distance between the occurrences of two events by specifying the minimum and maximum values in terms of a time granularity. When several constraints... more
Old-generation database models, such as the indexed-sequential, hierarchical, or network models, provide record-level access to their data, with all application logic residing in the hosting program. In contrast, relational databases can... more
In the haemodialysis domain, we are implementing a casebased architecture aimed at configuring Temporal Abstractions (TA) to be applied to time series data. The advantage of a case-based approach is the one of "quickly" obtaining a... more
Time series retrieval is a critical issue in all domains in which the observed phenomenon dynamics have to be dealt with. In this paper, we propose a novel, domain independent time series retrieval framework, based on Temporal... more
Interpreting time series of measurements and exploring a repository of cases with time series data looking for similarities, are nontrivial, but very important tasks. Classical methodological solutions proposed to deal with (some of)... more
In the hemodialysis domain, we are implementing a case-based, closed-loop architecture aimed at configuring temporal abstractions (TA), which will be applied to time series data. The advantage of a case-based approach is the one of... more
This paper introduces a framework and results of analyzing the hepatitis data with combination of temporal abstraction and data mining methods. In particular, we developed a novel temporal abstraction technique for temporal data... more
Temporal abstraction (TA) aims to transform temporal data into a symbolic interval-based representation of data. Most existing TA methods are applicable to data regularly collected in short periods. In this paper we proposed a TA method... more
The hepatitis temporal database collected at Chiba university hospital between 1982-2001 was recently given to challenge the KDD research. The database is large where each patient corresponds to 983 tests represented as sequences of... more
Classical stochastic Markov Decision Processes (MDPs) and possibilistic MDPs (-MDPs) aim at solving the same kind of problems, involving sequential decision making under uncertainty. The underlying uncertainty model (probabilistic /... more
In this work we propose a case-based architecture tackling the problem of configuring and processing temporal abstractions (trends and qualitative states) produced from raw time series data. The parameter configuration is a critical... more
Temporal abstraction (TA) aims to transform temporal data into a symbolic interval-based representation of data. Most existing TA methods are applicable to data regularly collected in short periods. In this paper we proposed a TA method... more
The hepatitis temporal database collected at Chiba university hospital between 1982-2001 was recently given to challenge the KDD research. The database is large where each patient corresponds to 983 tests represented as sequences of... more
We investigate the use of temporally abstract actions, or macro-actions, in the solution of Markov decision processes. Unlike current mod els that combine both primitive actions and macro-actions and leave the state space un changed, we... more
In this paper we investigate how timeless ontologies such as DFault, an ontology for fault diagnosis in power transmission networks can be re-engineered to include temporal entities. We propose a methodology, FONTE (Factorising ONTology... more
Records Denis Klimov and Yuval Shahar Medical Informatics Research Center Ben Gurion University of the Negev In a medical world with a large volume of time-stamped information, the clinicians and medical researchers need useful, intuitive... more
Therapy planning benefits from derived qualitative values or patterns which can be used for recommending therapeutic actions as well as for assessing the effectiveness of these actions within a certain period. Dealing with highfrequency... more
Sepsis is a condition caused by the body's overwhelming and life-threatening response to infection, which can lead to tissue damage, organ failure, and finally death. Common signs and symptoms include fever, increased heart rate,... more
Management of patients, especially chronic patients, requires presentation and processing of very large amounts of time-oriented clinical data. Using regular means such as text or tables is often ineffective, thus we propose to use the... more
The hepatitis temporal database collected at Chiba university hospital between 1982-2001 was recently given to challenge the KDD research. The database is large where each patient corresponds to 983 tests represented as sequences of... more
This paper contributes in two ways to the aims of this special issue on abstraction. The first is to show that there are compelling reasons motivating the use of abstraction in the purely computational realm of artificial intelligence.... more
Resource allocation is a difficult constraint sat isfaction problem that has many practical ap plications. Fully automatic systems are often rejected by the ultimate users because, in many real-world environments, constraints cannot be... more
The symposium grew out of a series of workshops on abstraction, approximation, and reformulation that had taken place alongside AAAI since 1989. This year's symposium was actually scheduled to take place at Lago Vista Clubs & Resort on... more
We describe a general method for abstracting higherlevel, interval-based concepts from time-stamped clinical data, the knowledge-based temporal-abstraction [KBTA] method. We focus on the knowledge representation, knowledge acquisition and... more
One of the fundamental challenges in reinforcement learning (RL) is the one of data efficiency: modern algorithms require a very large number of training samples, especially compared to humans, for solving environments with... more
Top-down processes are thought to play an important role in the mammalian visual system, e.g., for interpreting ambiguous stimuli. Slow Feature Analysis (SFA) [4] on the other hand is proven to be an efficient algorithm for the bottom-up... more
One of the fundamental challenges in reinforcement learning (RL) is the one of data efficiency: modern algorithms require a very large number of training samples, especially compared to humans, for solving environments with... more
In this work, we explore generative models based on temporally coherent representations. For this, we incorporate Slow Feature Analysis (SFA) into the encoder of a typical autoencoder architecture. We show that the latent factors... more
Studying the interactions between arbuscular mycorrhizal fungi (AMFs) and their symbiotic endobacteria has potentially strong impacts on the development of new biotechnology applications. The analysis of genomic data and syntenies is a... more
Background: Comparative genomics represents a key instrument to discover or validate phylogenetic relationships, to give insights on genome evolution, and to infer metabolic functions of a given organism. A tool for properly supporting... more
In this work we propose a case-based architecture tackling the problem of configuring and processing temporal abstractions (trends and qualitative states) produced from raw time series data. The parameter configuration is a critical... more
In the haemodialysis domain, we are implementing a casebased architecture aimed at configuring Temporal Abstractions (TA) to be applied to time series data. The advantage of a case-based approach is the one of "quickly" obtaining a... more
Studying arbuscular mycorrhizal fungi and their symbiotic endobacteria has potentially strong impacts on the development of new biotechnology applications. Comparative genomics is a key technique for acquiring information about... more
Studying the interactions between arbuscular mycorrhizal fungi (AMFs) and their symbiotic endobacteria has potentially strong impacts on the development of new biotechnology applications. The analysis of genomic data and syntenies is a... more
Time series retrieval is a critical issue in all domains in which the observed phenomenon dynamics have to be dealt with. In this paper, we propose a novel, domain independent time series retrieval framework, based on Temporal... more
Interpreting time series of measurements and exploring a repository of cases with time series data looking for similarities, are nontrivial, but very important tasks. Classical methodological solutions proposed to deal with (some of)... more
In the hemodialysis domain, we are implementing a case-based, closed-loop architecture aimed at configuring temporal abstractions (TA), which will be applied to time series data. The advantage of a case-based approach is the one of... more
Background: Comparative genomics represents a key instrument to discover or validate phylogenetic relationships, to give insights on genome evolution, and to infer metabolic functions of a given organism. A tool for properly supporting... more
There is a growing interest in intelligent assistants for a variety of applications from organizing tasks for knowledge workers to helping people with dementia. In our earlier work, we presented a decision-theoretic framework that... more