Sequential overchoice in product customization

The present study demonstrates how consumers can suffer from sequential overchoice. Customizing a tailor-made suit from combined-attribute choices (e.g., deciding on color and fabric in combination) leads to less satisfaction, more information overload, and less additional consumption than customizing it from single-attribute choices (e.g., deciding on color, then on fabric).

Corrected kriging update formulae for batch-sequential data assimilation

Recently, a lot of effort has been spent in the efficient computation of kriging predictors when observations are assimilated sequentially. In particular, kriging update formulae enabling significant computational savings were derived. Taking advantage of the previous kriging mean and variance computations helps avoiding a costly matrix inversion when adding one observation to the TeX already available ones. In addition to traditional update formulae taking into account a single new observation, Emery (2009) proposed formulae for the batch-sequential case, i.e. when TeX new observations are simultaneously assimilated. However, the kriging variance and covariance formulae given in Emery (2009) for the batch-sequential case are not correct. In this paper, we fix this issue and establish correct expressions for updated kriging variances and covariances when assimilating observations in parallel. An application in sequential conditional simulation finally shows that coupling update and residual substitution approaches may enable significant speed-ups.

KrigInv: An efficient and user-friendly implementation of batch-sequential inversion strategies based on kriging

Several strategies relying on kriging have recently been proposed for adaptively estimating contour lines and excursion sets of functions under severely limited evaluation budget. The recently released R package KrigInv 3 is presented and offers a sound implementation of various sampling criteria for those kinds of inverse problems. KrigInv is based on the DiceKriging package, and thus benefits from a number of options concerning the underlying kriging models. Six implemented sampling criteria are detailed in a tutorial and illustrated with graphical examples. Different functionalities of KrigInv are gradually explained. Additionally, two recently proposed criteria for batch-sequential inversion are presented, enabling advanced users to distribute function evaluations in parallel on clusters or clouds of machines. Finally, auxiliary problems are discussed. These include the fine tuning of numerical integration and optimization procedures used within the computation and the optimization of the considered criteria.

Repeatedly experiencing information overload in a product customization: a case of sequential overchoice

The present study demonstrates how consumers can suffer from sequential overchoice. Customizing a tailor-made suit from combined-attribute choices (e.g., deciding on color and fabric in combination) leads to less satisfaction and less additional consumption than customizing it from single-attribute choices (e.g., deciding on color, then on fabric). The effect is mediated by information overload and moderated by consideration set size.

Quantifying uncertainty on Pareto fronts with Gaussian process conditional simulations

Multi-objective optimization algorithms aim at finding Pareto-optimal solutions. Recovering Pareto fronts or Pareto sets from a limited number of function evaluations are challenging problems. A popular approach in the case of expensive-to-evaluate functions is to appeal to metamodels. Kriging has been shown efficient as a base for sequential multi-objective optimization, notably through infill sampling criteria balancing exploitation and exploration such as the Expected Hypervolume Improvement. Here we consider Kriging metamodels not only for selecting new points, but as a tool for estimating the whole Pareto front and quantifying how much uncertainty remains on it at any stage of Kriging-based multi-objective optimization algorithms. Our approach relies on the Gaussian process interpretation of Kriging, and bases upon conditional simulations. Using concepts from random set theory, we propose to adapt the Vorob’ev expectation and deviation to capture the variability of the set of non-dominated points. Numerical experiments illustrate the potential of the proposed workflow, and it is shown on examples how Gaussian process simulations and the estimated Vorob’ev deviation can be used to monitor the ability of Kriging-based multi-objective optimization algorithms to accurately learn the Pareto front.

Necessity of epicardial ablation for ventricular tachycardia after sequential endocardial approach

Background Catheter ablation (CA) of ventricular tachycardia (VT) is an important treatment option in patients with structural heart disease (SHD) and implantable cardioverter defibrillator (ICD). A subset of patients requires epicardial CA for VT. Objective The purpose of the study was to assess the significance of epicardial CA in these patients after a systematic sequential endocardial approach. Methods Between January 2009 and October 2012 CA for VT was analyzed. A sequential CA approach guided by earliest ventricular activation, pacemap, entrainment and stimulus to QRS-interval analysis was used. Acute CA success was assessed by programmed ventricular stimulation. ICD interrogation and 24 h-Holter ECG were used to evaluate long-term success. Results One hundred sixty VT ablation procedures in 126 consecutive patients (114 men; age 65 ± 12 years) were performed. Endocardial CA succeeded in 250 (94%) out of 265 treated VT. For 15 (6%) VT an additional epicardial CA was performed and succeeded in 9 of these 15 VT. Long-term FU (25 ± 18.2 month) showed freedom of VT in 104 pts (82%) after 1.2 ± 0.5 procedures, 11 (9%) suffered from repeated ICD shocks and 11 (9%) died due to worsening of heart failure. Conclusions Despite a heterogenic substrate for VT in SHD, endocardial CA alone results in high acute success rates. In this study additional epicardial CA following a sequential endocardial mapping and CA approach was performed in 6% of VT. Thus, due to possible complications epicardial CA should only be considered if endocardial CA fails.

The effectiveness and safety of rescue treatments in 108 patients with steroid-refractory ulcerative colitis with sequential rescue therapies in a subgroup of patients

BACKGROUND: Among patients with steroid-refractory ulcerative colitis (UC) in whom a first rescue therapy has failed, a second line salvage treatment can be considered to avoid colectomy. AIM: To evaluate the efficacy and safety of second or third line rescue therapy over a one-year period. METHODS: Response to single or sequential rescue treatments with infliximab (5mg/kg intravenously (iv) at week 0, 2, 6 and then every 8weeks), ciclosporin (iv 2mg/kg/daily and then oral 5mg/kg/daily) or tacrolimus (0.05mg/kg divided in 2 doses) in steroid-refractory moderate to severe UC patients from 7 Swiss and 1 Serbian tertiary IBD centers was retrospectively studied. The primary endpoint was the one year colectomy rate. RESULTS: 60% of patients responded to the first rescue therapy, 10% went to colectomy and 30% non-responders were switched to a 2(nd) line rescue treatment. 66% of patients responded to the 2(nd) line treatment whereas 34% failed, of which 15% went to colectomy and 19% received a 3(rd) line rescue treatment. Among those, 50% patients went to colectomy. Overall colectomy rate of the whole cohort was 18%. Steroid-free remission rate was 39%. The adverse event rates were 33%, 37.5% and 30% for the first, second and third line treatment respectively. CONCLUSION: Our data show that medical intervention even with 2(nd) and 3(rd) rescue treatments decreased colectomy frequency within one year of follow up. A longer follow-up will be necessary to investigate whether sequential therapy will only postpone colectomy and what percentage of patients will remain in long-term remission.

The Exploitation of Versatile Building Blocks for the Self-Assembly of Novel Molecular Magnets

Using molecular building blocks to self-assemble lattices supporting long-range magnetic order is currently an active area of solid-state chemistry. Consequently, it is the realm of supramolecular chemistry that synthetic chemists are turning to in order to develop techniques for the synthesis of structurally well-defined supramolecular materials. In recent years we have investigated the versatility and usefulness of two classes of molecular building blocks, namely, tris-oxalato transition-metal (M. Pilkington and S. Decurtins, in “Magnetoscience—From Molecules to Materials,” Wiley–VCH, 2000), and octacyanometalate complexes (Pilkington and Decurtins, Chimia 54, 593 (2001)), for applications in the field of molecule-based magnets. Anionic, tris-chelated oxalato building blocks are able to build up two-dimensional honeycomb-layered structural motifs as well as three-dimensional decagon frameworks. The discrimination between the crystallization of the two- or three-dimensional structures relies on the choice of the templating counterions (Decurtins, Chimia 52, 539 (1998); Decurtins et al. Mol. Cryst. Liq. Cryst. 273, 167 (1995); New J. Chem. 117 (1998)). These structural types display a range of ferro, ferri, and antiferromagnetic properties (Pilkington and Decurtins, in “Magnetoscience—From Molecules to Materials”). Octacyanometalate building blocks self-assemble to afford two new classes of cyano-bridged compounds namely, molecular clusters and extended three dimensional networks (J. Larionova et al., Angew. Chem. Int. Ed. 39, 1605 (2000); Pilkington et al., in preparation). The molecular cluster with a MnII9MoV6 core has the highest ground state spin value, S=51/2, reported to-date (Larionova et al., Angew. Chem. Int. Ed. 39, 1605 (2000)). In the high-temperature regime, the magnetic properties are characterized by ferromagnetic intracluster coupling. In the magnetic range below 44 K, the magnetic cluster signature is lost as possibly a bulk behavior starts to emerge. The three-dimensional networks exhibit both paramagnetic and ferromagnetic behavior, since the magnetic properties of these materials directly reflect the electronic configuration of the metal ion incorporated into the octacyanometalate building blocks (Pilkington et al., in preparation). For both the oxalate- and cyanide-bridged materials, we are able to manipulate the magnetic properties of the supramolecular assemblies by tuning the electronic configurations of the metal ions incorporated into the appropriate molecular building blocks (Pilkington and Decurtins, in “Magnetoscience—From Molecules to Materials,” Chimia 54, 593 (2000)).