LJLuc Jaulmes

Luc Jaulmes

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Luc Jaulmes has a total of 26 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Tensorize: Fast Synthesis of Tensor Programs from Legacy Code using Symbolic Tracing, Sketching and Solving

    by Alexander Brauckmann, Luc Jaulmes, José W. de Souza Magalhães, Elizabeth Polgreen, Michael F. P. O’bOyle

    Tensor domain specific languages (DSLs) achieve substantial performance due to high-level compiler optimization and hardware acceleration. However, to achieve such performance for existing applications requires the programmer to manual rewrite their legacy code in evolving Tensor DSLs. Prior efforts to automate this translation face significant scalability issues which greatly reduces their applicability to real-world code. This paper presents Tensorize, a novel MLIR-based compiler approach to automatically lift legacy code to high level Tensor DSLs using program synthesis. Tensorizeuses a symbolic trace of the legacy program as a specification and automatically selects sketches from the target Tensor DSLs to drive the program synthesis. It uses an algebraic solver to rapidly simplify the specification, resulting in a fast, automatic approach that is correct by design. We evaluate Tensorizeon several legacy code benchmarks and compare against state-of-the-art techniques. Tensorizeis able to lift more code than prior schemes, is an order of magnitude faster in synthesis time, and guarantees correctness by construction.


  • Open Access

    Effectiveness and Medicoeconomic Evaluation of Home Monitoring of Patients With Mild COVID-19: Covidom Cohort Study

    by Luc Jaulmes, Youri Yordanov, Alexandre Descamps, Isabelle Durand-Zaleski, Aurélien Dinh, Patrick Jourdain, Agnès Dechartres

    Background: Covidom was a telemonitoring solution for home monitoring of patients with mild to moderate COVID-19, deployed in March 2020 in the Greater Paris area in France to alleviate the burden on the health care system. The Covidom solution included a free mobile application with daily monitoring questionnaires and a regional control center to quickly handle patient alerts, including dispatching emergency medical services when necessary. Objective: This study aimed to provide an overall evaluation of the Covidom solution 18 months after its inception in terms of effectiveness, safety, and cost. Methods: Our primary outcome was to measure effectiveness using the number of handled alerts, response escalation, and patient-reported medical contacts outside of Covidom. Then, we analyzed the safety of Covidom by assessing its ability to detect clinical worsening, defined as hospitalization or death, and the number of patients with clinical worsening without any preceding alert. We evaluated the cost of Covidom and compared the cost of hospitalization for Covidom and non-Covidom patients with mild COVID-19 cases seen in the emergency departments of the largest network of hospitals in the Greater Paris area (Assistance Publique-Hôpitaux de Paris). Finally, we reported on user satisfaction. Results: Of the 60,073 patients monitored by Covidom, the regional control center handled 285,496 alerts and dispatched emergency medical services 518 times. Of the 13,204 respondents who responded to either of the follow-up questionnaires, 65.8% (n=8690) reported having sought medical care outside the Covidom solution during their monitoring period. Of the 947 patients who experienced clinical worsening while adhering to daily monitoring, only 35 (3.7%) did not previously trigger alerts (35 were hospitalized, including 1 who died). The average cost of Covidom was €54 (US $1=€0.8614) per patient, and the cost of hospitalization for COVID-19 worsening was significantly lower in Covidom than in non-Covidom patients with mild COVID-19 cases seen in the emergency departments of Assistance Publique-Hôpitaux de Paris. The patients who responded to the satisfaction questionnaire had a median rating of 9 (out of 10) for the likelihood of recommending Covidom. Conclusions: Covidom may have contributed to alleviating the pressure on the health care system in the initial months of the pandemic, although its impact was lower than anticipated, with a substantial number of patients having consulted outside of Covidom. Covidom seems to be safe for home monitoring of patients with mild to moderate COVID-19.


  • Open Access

    Effectiveness and Medicoeconomic Evaluation of Home Monitoring of Patients With Mild COVID-19: Covidom Cohort Study (Preprint)

    by Luc Jaulmes, Youri Yordanov, Alexandre Descamps, Isabelle Durand-Zaleski, Aurélien Dinh, Patrick Jourdain, Agnès Dechartres

    BACKGROUND Covidom was a telemonitoring solution for home monitoring of patients with mild to moderate COVID-19, deployed in March 2020 in the Greater Paris area in France to alleviate the burden on the health care system. The Covidom solution included a free mobile application with daily monitoring questionnaires and a regional control center to quickly handle patient alerts, including dispatching emergency medical services when necessary. OBJECTIVE This study aimed to provide an overall evaluation of the Covidom solution 18 months after its inception in terms of effectiveness, safety, and cost. METHODS Our primary outcome was to measure effectiveness using the number of handled alerts, response escalation, and patient-reported medical contacts outside of Covidom. Then, we analyzed the safety of Covidom by assessing its ability to detect clinical worsening, defined as hospitalization or death, and the number of patients with clinical worsening without any preceding alert. We evaluated the cost of Covidom and compared the cost of hospitalization for Covidom and non-Covidom patients with mild COVID-19 cases seen in the emergency departments of the largest network of hospitals in the Greater Paris area (Assistance Publique-Hôpitaux de Paris). Finally, we reported on user satisfaction. RESULTS Of the 60,073 patients monitored by Covidom, the regional control center handled 285,496 alerts and dispatched emergency medical services 518 times. Of the 13,204 respondents who responded to either of the follow-up questionnaires, 65.8% (n=8690) reported having sought medical care outside the Covidom solution during their monitoring period. Of the 947 patients who experienced clinical worsening while adhering to daily monitoring, only 35 (3.7%) did not previously trigger alerts (35 were hospitalized, including 1 who died). The average cost of Covidom was €54 (US $1=€0.8614) per patient, and the cost of hospitalization for COVID-19 worsening was significantly lower in Covidom than in non-Covidom patients with mild COVID-19 cases seen in the emergency departments of Assistance Publique-Hôpitaux de Paris. The patients who responded to the satisfaction questionnaire had a median rating of 9 (out of 10) for the likelihood of recommending Covidom. CONCLUSIONS Covidom may have contributed to alleviating the pressure on the health care system in the initial months of the pandemic, although its impact was lower than anticipated, with a substantial number of patients having consulted outside of Covidom. Covidom seems to be safe for home monitoring of patients with mild to moderate COVID-19.

  • Open Access

    Characteristics and outcomes of asthmatic outpatients with COVID-19 who receive home telesurveillance

    by Antoine Beurnier, Youri Yordanov, Agnès Dechartres, Aurélien Dinh, Erwan Debuc, François-Xavier Lescure, Patrick Jourdain, Luc Jaulmes, Marc Humbert

    Background: The prognosis of asthmatic outpatients with COVID-19 needs to be clarified. The objectives of this study were: 1) to investigate the characteristics and outcomes of asthmatic patients receiving initial ambulatory care and home monitoring for COVID-19 with Covidom, a telesurveillance solution; and 2) to compare the characteristics and outcomes between asthmatic and non-asthmatic patients.Methods: Inclusion criteria were age ≥18 years, suspected or confirmed COVID-19 diagnosis allowing initial ambulatory care, registration in Covidom between March 2020 and April 2021 and completion of the initial medical questionnaire. We compared clinical characteristics and outcomes between asthmatic and non-asthmatic patients, and we evaluated whether asthma was independently associated with clinical worsening (hospitalisation or death) within 30 days follow-up using a multivariate logistic regression model.Results: 33 815 patients met the inclusion criteria. Asthma was reported in 4276 (12.6%). The main comorbidities among asthmatic patients were obesity (23.1%), hypertension (12.7%) and diabetes (4.5%). As compared with non-asthmatic patients, asthmatic patients were more often female (70.0versus62.1%, pversus43.8 years, pversus17.6%, pversus4.2%, p=0.203) and no asthmatic patient died during follow-up (versus25 non-asthmatic patients, 0.1%; p=0.109). In multivariate analysis, asthma was independently associated with higher risk of clinical worsening (OR 1.23, 95% CI 1.04–1.44, p=0.013).Conclusion: In a large French cohort of patients receiving initial ambulatory care and home monitoring for COVID-19, asthma was independently associated with higher risk of clinical worsening although no asthmatic patient died within the 30 days follow-up.


  • Open Access

    Memory Vulnerability: A Case for Delaying Error Reporting

    by Luc Jaulmes, Miquel Moretó, Mateo Valero, Marc Casas

    To face future reliability challenges, it is necessary to quantify the risk of error in any part of a computing system. To this goal, the Architectural Vulnerability Factor (AVF) has long been used for chips. However, this metric is used for offline characterisation, which is inappropriate for memory. We survey the literature and formalise one of the metrics used, the Memory Vulnerability Factor, and extend it to take into account false errors. These are reported errors which would have no impact on the program if they were ignored. We measure the False Error Aware MVF (FEA) and related metrics precisely in a cycle-accurate simulator, and compare them with the effects of injecting faults in a program's data, in native parallel runs. Our findings show that MVF and FEA are the only two metrics that are safe to use at runtime, as they both consistently give an upper bound on the probability of incorrect program outcome. FEA gives a tighter bound than MVF, and is the metric that correlates best with the incorrect outcome probability of all considered metrics.

  • Asynchronous and Exact Forward Recovery for Detected Errors in Iterative Solvers

    by Luc Jaulmes, Miquel Moreto, Eduard Ayguade, Jesus Labarta, Mateo Valero, Marc Casas

    Current trends and projections show that faults in computer systems become increasingly common. Such errors may be detected, and possibly corrected transparently, e.g. by Error Correcting Codes (ECC). For a program to be fault-tolerant, it needs to also handle the Errors that are Detected and Uncorrected (DUE), such as an ECC encountering too many bit flips in a codeword. While correcting an error has an overhead in itself, it can also affect the progress of a program. The most generic technique, rolling back the program state to a previously taken checkpoint, sets back any progress done since then. Alternately, application specific techniques exist, such as restarting an iterative program with its latest iteration's values as initial guess.