@inproceedings{1859, abstract = {Structural support vector machines (SSVMs) are amongst the best performing models for structured computer vision tasks, such as semantic image segmentation or human pose estimation. Training SSVMs, however, is computationally costly, because it requires repeated calls to a structured prediction subroutine (called \emph{max-oracle}), which has to solve an optimization problem itself, e.g. a graph cut. In this work, we introduce a new algorithm for SSVM training that is more efficient than earlier techniques when the max-oracle is computationally expensive, as it is frequently the case in computer vision tasks. The main idea is to (i) combine the recent stochastic Block-Coordinate Frank-Wolfe algorithm with efficient hyperplane caching, and (ii) use an automatic selection rule for deciding whether to call the exact max-oracle or to rely on an approximate one based on the cached hyperplanes. We show experimentally that this strategy leads to faster convergence to the optimum with respect to the number of requires oracle calls, and that this translates into faster convergence with respect to the total runtime when the max-oracle is slow compared to the other steps of the algorithm. }, author = {Shah, Neel and Kolmogorov, Vladimir and Lampert, Christoph}, location = {Boston, MA, USA}, pages = {2737 -- 2745}, publisher = {IEEE}, title = {{A multi-plane block-coordinate Frank-Wolfe algorithm for training structural SVMs with a costly max-oracle}}, doi = {10.1109/CVPR.2015.7298890}, year = {2015}, } @inproceedings{1860, abstract = {Classifiers for object categorization are usually evaluated by their accuracy on a set of i.i.d. test examples. This provides us with an estimate of the expected error when applying the classifiers to a single new image. In real application, however, classifiers are rarely only used for a single image and then discarded. Instead, they are applied sequentially to many images, and these are typically not i.i.d. samples from a fixed data distribution, but they carry dependencies and their class distribution varies over time. In this work, we argue that the phenomenon of correlated data at prediction time is not a nuisance, but a blessing in disguise. We describe a probabilistic method for adapting classifiers at prediction time without having to retrain them. We also introduce a framework for creating realistically distributed image sequences, which offers a way to benchmark classifier adaptation methods, such as the one we propose. Experiments on the ILSVRC2010 and ILSVRC2012 datasets show that adapting object classification systems at prediction time can significantly reduce their error rate, even with no additional human feedback.}, author = {Royer, Amélie and Lampert, Christoph}, location = {Boston, MA, United States}, pages = {1401 -- 1409}, publisher = {IEEE}, title = {{Classifier adaptation at prediction time}}, doi = {10.1109/CVPR.2015.7298746}, year = {2015}, } @article{1861, abstract = {Continuous-time Markov chains are commonly used in practice for modeling biochemical reaction networks in which the inherent randomness of themolecular interactions cannot be ignored. This has motivated recent research effort into methods for parameter inference and experiment design for such models. The major difficulty is that such methods usually require one to iteratively solve the chemical master equation that governs the time evolution of the probability distribution of the system. This, however, is rarely possible, and even approximation techniques remain limited to relatively small and simple systems. An alternative explored in this article is to base methods on only some low-order moments of the entire probability distribution. We summarize the theory behind such moment-based methods for parameter inference and experiment design and provide new case studies where we investigate their performance.}, author = {Ruess, Jakob and Lygeros, John}, journal = {ACM Transactions on Modeling and Computer Simulation}, number = {2}, publisher = {ACM}, title = {{Moment-based methods for parameter inference and experiment design for stochastic biochemical reaction networks}}, doi = {10.1145/2688906}, volume = {25}, year = {2015}, } @article{1864, abstract = {The Altshuler–Shklovskii formulas (Altshuler and Shklovskii, BZh Eksp Teor Fiz 91:200, 1986) predict, for any disordered quantum system in the diffusive regime, a universal power law behaviour for the correlation functions of the mesoscopic eigenvalue density. In this paper and its companion (Erdős and Knowles, The Altshuler–Shklovskii formulas for random band matrices I: the unimodular case, 2013), we prove these formulas for random band matrices. In (Erdős and Knowles, The Altshuler–Shklovskii formulas for random band matrices I: the unimodular case, 2013) we introduced a diagrammatic approach and presented robust estimates on general diagrams under certain simplifying assumptions. In this paper, we remove these assumptions by giving a general estimate of the subleading diagrams. We also give a precise analysis of the leading diagrams which give rise to the Altschuler–Shklovskii power laws. Moreover, we introduce a family of general random band matrices which interpolates between real symmetric (β = 1) and complex Hermitian (β = 2) models, and track the transition for the mesoscopic density–density correlation. Finally, we address the higher-order correlation functions by proving that they behave asymptotically according to a Gaussian process whose covariance is given by the Altshuler–Shklovskii formulas. }, author = {Erdös, László and Knowles, Antti}, journal = {Annales Henri Poincare}, number = {3}, pages = {709 -- 799}, publisher = {Springer}, title = {{The Altshuler-Shklovskii formulas for random band matrices II: The general case}}, doi = {10.1007/s00023-014-0333-5}, volume = {16}, year = {2015}, } @article{1865, abstract = {The plant hormone auxin and its directional transport are known to play a crucial role in defining the embryonic axis and subsequent development of the body plan. Although the role of PIN auxin efflux transporters has been clearly assigned during embryonic shoot and root specification, the role of the auxin influx carriers AUX1 and LIKE-AUX1 (LAX) proteins is not well established. Here, we used chemical and genetic tools on Brassica napus microspore-derived embryos and Arabidopsis thaliana zygotic embryos, and demonstrate that AUX1, LAX1 and LAX2 are required for both shoot and root pole formation, in concert with PIN efflux carriers. Furthermore, we uncovered a positive-feedback loop betweenMONOPTEROS(ARF5)-dependent auxin signalling and auxin transport. ThisMONOPTEROSdependent transcriptional regulation of auxin influx (AUX1, LAX1 and LAX2) and auxin efflux (PIN1 and PIN4) carriers by MONOPTEROS helps to maintain proper auxin transport to the root tip. These results indicate that auxin-dependent cell specification during embryo development requires balanced auxin transport involving both influx and efflux mechanisms, and that this transport is maintained by a positive transcriptional feedback on auxin signalling.}, author = {Robert, Hélène and Grunewald, Wim and Sauer, Michael and Cannoot, Bernard and Soriano, Mercedes and Swarup, Ranjan and Weijers, Dolf and Bennett, Malcolm and Boutilier, Kim and Friml, Jirí}, journal = {Development}, number = {4}, pages = {702 -- 711}, publisher = {Company of Biologists}, title = {{Plant embryogenesis requires AUX/LAX-mediated auxin influx}}, doi = {10.1242/dev.115832}, volume = {142}, year = {2015}, } @article{1866, author = {Henzinger, Thomas A and Raskin, Jean}, journal = {Communications of the ACM}, number = {2}, pages = {86--86}, publisher = {ACM}, title = {{The equivalence problem for finite automata: Technical perspective}}, doi = {10.1145/2701001}, volume = {58}, year = {2015}, } @article{1867, abstract = {Cultured mammalian cells essential are model systems in basic biology research, production platforms of proteins for medical use, and testbeds in synthetic biology. Flavin cofactors, in particular flavin mononucleotide (FMN) and flavin adenine dinucleotide (FAD), are critical for cellular redox reactions and sense light in naturally occurring photoreceptors and optogenetic tools. Here, we quantified flavin contents of commonly used mammalian cell lines. We first compared three procedures for extraction of free and noncovalently protein-bound flavins and verified extraction using fluorescence spectroscopy. For separation, two CE methods with different BGEs were established, and detection was performed by LED-induced fluorescence with limit of detections (LODs 0.5-3.8 nM). We found that riboflavin (RF), FMN, and FAD contents varied significantly between cell lines. RF (3.1-14 amol/cell) and FAD (2.2-17.0 amol/cell) were the predominant flavins, while FMN (0.46-3.4 amol/cell) was found at markedly lower levels. Observed flavin contents agree with those previously extracted from mammalian tissues, yet reduced forms of RF were detected that were not described previously. Quantification of flavins in mammalian cell lines will allow a better understanding of cellular redox reactions and optogenetic tools.}, author = {Hühner, Jens and Inglés Prieto, Álvaro and Neusüß, Christian and Lämmerhofer, Michael and Janovjak, Harald L}, journal = {Electrophoresis}, number = {4}, pages = {518 -- 525}, publisher = {Wiley}, title = {{Quantification of riboflavin, flavin mononucleotide, and flavin adenine dinucleotide in mammalian model cells by CE with LED-induced fluorescence detection}}, doi = {10.1002/elps.201400451}, volume = {36}, year = {2015}, } @article{1868, abstract = {We investigate high-dimensional nonlinear dynamical systems exhibiting multiple resonances under adiabatic parameter variations. Our motivations come from experimental considerations where time-dependent sweeping of parameters is a practical approach to probing and characterizing the bifurcations of the system. The question is whether bifurcations so detected are faithful representations of the bifurcations intrinsic to the original stationary system. Utilizing a harmonically forced, closed fluid flow system that possesses multiple resonances and solving the Navier-Stokes equation under proper boundary conditions, we uncover the phenomenon of the early effect. Specifically, as a control parameter, e.g., the driving frequency, is adiabatically increased from an initial value, resonances emerge at frequency values that are lower than those in the corresponding stationary system. The phenomenon is established by numerical characterization of physical quantities through the resonances, which include the kinetic energy and the vorticity field, and a heuristic analysis based on the concept of instantaneous frequency. A simple formula is obtained which relates the resonance points in the time-dependent and time-independent systems. Our findings suggest that, in general, any true bifurcation of a nonlinear dynamical system can be unequivocally uncovered through adiabatic parameter sweeping, in spite of a shift in the bifurcation point, which is of value to experimental studies of nonlinear dynamical systems.}, author = {Park, Youngyong and Do, Younghae and Altmeyer, Sebastian and Lai, Yingcheng and Lee, Gyuwon}, issn = {1539-3755}, journal = {Physical Review E}, number = {2}, publisher = {American Physical Society}, title = {{Early effect in time-dependent, high-dimensional nonlinear dynamical systems with multiple resonances}}, doi = {10.1103/PhysRevE.91.022906}, volume = {91}, year = {2015}, } @article{1871, abstract = {The plant hormone auxin is a key regulator of plant growth and development. Differences in auxin distribution within tissues are mediated by the polar auxin transport machinery, and cellular auxin responses occur depending on changes in cellular auxin levels. Multiple receptor systems at the cell surface and in the interior operate to sense and interpret fluctuations in auxin distribution that occur during plant development. Until now, three proteins or protein complexes that can bind auxin have been identified. SCFTIR1 [a SKP1-cullin-1-F-box complex that contains transport inhibitor response 1 (TIR1) as the F-box protein] and S-phase-kinaseassociated protein 2 (SKP2) localize to the nucleus, whereas auxinbinding protein 1 (ABP1), predominantly associates with the endoplasmic reticulum and cell surface. In this Cell Science at a Glance article, we summarize recent discoveries in the field of auxin transport and signaling that have led to the identification of new components of these pathways, as well as their mutual interaction.}, author = {Grones, Peter and Friml, Jirí}, journal = {Journal of Cell Science}, number = {1}, pages = {1 -- 7}, publisher = {Company of Biologists}, title = {{Auxin transporters and binding proteins at a glance}}, doi = {10.1242/jcs.159418}, volume = {128}, year = {2015}, } @article{1873, abstract = {We consider partially observable Markov decision processes (POMDPs) with limit-average payoff, where a reward value in the interval [0,1] is associated with every transition, and the payoff of an infinite path is the long-run average of the rewards. We consider two types of path constraints: (i) a quantitative constraint defines the set of paths where the payoff is at least a given threshold λ1ε(0,1]; and (ii) a qualitative constraint which is a special case of the quantitative constraint with λ1=1. We consider the computation of the almost-sure winning set, where the controller needs to ensure that the path constraint is satisfied with probability 1. Our main results for qualitative path constraints are as follows: (i) the problem of deciding the existence of a finite-memory controller is EXPTIME-complete; and (ii) the problem of deciding the existence of an infinite-memory controller is undecidable. For quantitative path constraints we show that the problem of deciding the existence of a finite-memory controller is undecidable. We also present a prototype implementation of our EXPTIME algorithm and experimental results on several examples.}, author = {Chatterjee, Krishnendu and Chmelik, Martin}, journal = {Artificial Intelligence}, pages = {46 -- 72}, publisher = {Elsevier}, title = {{POMDPs under probabilistic semantics}}, doi = {10.1016/j.artint.2014.12.009}, volume = {221}, year = {2015}, } @inbook{18734, abstract = {In this Introduction, we outline expectations for when and how the hydrogen and helium atoms in the universe turned from neutral to ionized, focusing on the earliest, least well understood stages, and emphasize the most important open questions. We include a historical summary, and highlight the role of reionization as one of the few milestones in the evolution of the universe since the Big Bang, and its status as a unique probe of the beginning stages of structure formation.}, author = {Haiman, Zoltán}, booktitle = {Understanding the Epoch of Cosmic Reionization}, editor = {Mesinger, Andrei}, isbn = {9783319219561}, issn = {2214-7985}, pages = {1--22}, publisher = {Springer Nature}, title = {{Cosmic Reionization and the First Nonlinear Structures in the Universe}}, doi = {10.1007/978-3-319-21957-8_1}, year = {2015}, } @article{12626, abstract = {Ice cliffs have been identified as a reason for higher ablation rates on debris-covered glaciers than are implied by the insulation effects of the debris. This study aims to improve our understanding of cliff backwasting, and the role of radiative fluxes in particular. An energy-balance model is forced with new data gathered in May and October 2013 on Lirung Glacier, Nepalese Himalaya. Observations show substantial variability in melt between cliffs, between locations on any cliff and between seasons. Using a high-resolution digital elevation model we calculate longwave fluxes incident to the cliff from surrounding terrain and include the effect of local shading on shortwave radiation. This is an advance over previous studies, that made simplified assumptions on cliff geometry and radiative fluxes. Measured melt rates varied between 3.25 and 8.6 cm d−1 in May and 0.18 and 1.34 cm d−1 in October. Model results reproduce the strong variability in space and time, suggesting considerable differences in radiative fluxes over one cliff. In October the model fails to reproduce stake readings, probably due to the lack of a refreezing component. Disregarding local topography can lead to overestimation of melt at the point scale by up to ∼9%.}, author = {Steiner, Jakob F. and Pellicciotti, Francesca and Buri, Pascal and Miles, Evan S. and Immerzeel, Walter W. and Reid, Tim D.}, issn = {1727-5652}, journal = {Journal of Glaciology}, number = {229}, pages = {889--907}, publisher = {International Glaciological Society}, title = {{Modelling ice-cliff backwasting on a debris-covered glacier in the Nepalese Himalaya}}, doi = {10.3189/2015jog14j194}, volume = {61}, year = {2015}, } @article{12627, abstract = {Spatial evolution of supraglacial debris cover on mountain glaciers is a largely unmonitored and poorly understood phenomenon that directly affects glacier melt. Supraglacial debris cover for 93 glaciers in the Karakoram, northern Pakistan, was mapped from Landsat imagery acquired in 1977, 1998, 2009 and 2014. Surge-type glaciers occupy 41% of the study area and were considered separately. The time series of debris-covered surface area change shows a mean value of zero or near-zero change for both surging and non-surging glaciers. An increase in debris-covered area is often associated with negative regional mass balances. We extend this logic to suggest that the stable regional mass balances in the Karakoram explain the zero or near-zero change in debris-covered area. This coupling of trends combined with our 37 year time series of data suggests the Karakoram anomaly extends further back in time than previously known.}, author = {Herreid, Sam and Pellicciotti, Francesca and Ayala, Alvaro and Chesnokova, Anna and Kienholz, Christian and Shea, Joseph and Shrestha, Arun}, issn = {1727-5652}, journal = {Journal of Glaciology}, number = {227}, pages = {524--536}, publisher = {International Glaciological Society}, title = {{Satellite observations show no net change in the percentage of supraglacial debris-covered area in northern Pakistan from 1977 to 2014}}, doi = {10.3189/2015jog14j227}, volume = {61}, year = {2015}, } @article{12628, abstract = {Thick debris cover on glaciers can significantly reduce ice melt. However, several studies have suggested that debris-covered glaciers in the Himalaya might have lost mass at a rate similar to debris-free glaciers. We reconstruct elevation and mass changes for the debris-covered glaciers of the upper Langtang valley, Nepalese Himalaya, using a digital elevation model (DEM) from 1974 stereo Hexagon satellite data and the 2000 SRTM (Shuttle Radar Topography Mission) DEM. Uncertainties are high in the accumulation areas, due to data gaps in the SRTM and difficulties with delineation of the glacier borders. Even with these uncertainties, we obtain thinning rates comparable to those of several other studies in the Himalaya. In particular, we obtain a total mass balance for the investigated debris-covered glaciers of the basin of –0.32 ± 0.18 m w.e. a−1. However, there are major spatial differences both between glaciers and within any single glacier, exhibiting a very distinct nonlinear mass-balance profile with elevation. Through analysis of surface velocities derived from Landsat ETM+ imagery, we show that thinning occurs in areas of low velocity and low slope. These areas are prone to a general, dynamic decay of surface features and to the development of supraglacial lakes and ice cliffs, which may be responsible for a considerable increase in overall glacier ablation.}, author = {Pellicciotti, Francesca and Stephan, Christa and Miles, Evan and Herreid, Sam and Immerzeel, Walter W. and Bolch, Tobias}, issn = {1727-5652}, journal = {Journal of Glaciology}, keywords = {Earth-Surface Processes}, number = {226}, pages = {373--386}, publisher = {International Glaciological Society}, title = {{Mass-balance changes of the debris-covered glaciers in the Langtang Himal, Nepal, from 1974 to 1999}}, doi = {10.3189/2015jog13j237}, volume = {61}, year = {2015}, } @article{12629, abstract = {Meteorological studies in high-mountain environments form the basis of our understanding of catchment hydrology and glacier accumulation and melt processes, yet high-altitude (>4000 m above sea level, asl) observatories are rare. This research presents meteorological data recorded between December 2012 and November 2013 at seven stations in Nepal, ranging in elevation from 3860 to 5360 m asl. Seasonal and diurnal cycles in air temperature, vapour pressure, incoming short-wave and long-wave radiation, atmospheric transmissivity, wind speed, and precipitation are compared between sites. Solar radiation strongly affects diurnal temperature and vapour pressure cycles, but local topography and valley-scale circulations alter wind speed and precipitation cycles. The observed diurnal variability in vertical temperature gradients in all seasons highlights the importance of in situ measurements for melt modelling. The monsoon signal (progressive onset and sharp end) is visible in all data-sets, and the passage of the remnants of Typhoon Phailin in mid-October 2013 provides an interesting case study on the possible effects of such storms on glaciers in the region.}, author = {Shea, J.M. and Wagnon, P. and Immerzeel, W.W. and Biron, R. and Brun, F. and Pellicciotti, Francesca}, issn = {1360-0648}, journal = {International Journal of Water Resources Development}, keywords = {Water Science and Technology, Development}, number = {2}, pages = {174--200}, publisher = {Taylor & Francis}, title = {{A comparative high-altitude meteorological analysis from three catchments in the Nepalese Himalaya}}, doi = {10.1080/07900627.2015.1020417}, volume = {31}, year = {2015}, } @article{12630, abstract = {The hydrology of high-elevation watersheds of the Hindu Kush-Himalaya region (HKH) is poorly known. The correct representation of internal states and process dynamics in glacio-hydrological models can often not be verified due to missing in situ measurements. We use a new set of detailed ground data from the upper Langtang valley in Nepal to systematically guide a state-of-the art glacio-hydrological model through a parameter assigning process with the aim to understand the hydrology of the catchment and contribution of snow and ice processes to runoff. 14 parameters are directly calculated on the basis of local data, and 13 parameters are calibrated against 5 different datasets of in situ or remote sensing data. Spatial fields of debris thickness are reconstructed through a novel approach that employs data from an Unmanned Aerial Vehicle (UAV), energy balance modeling and statistical techniques. The model is validated against measured catchment runoff (Nash–Sutcliffe efficiency 0.87) and modeled snow cover is compared to Landsat snow cover. The advanced representation of processes allowed assessing the role played by avalanching for runoff for the first time for a Himalayan catchment (5% of annual water inputs to the hydrological system are due to snow redistribution) and to quantify the hydrological significance of sub-debris ice melt (9% of annual water inputs). Snowmelt is the most important contributor to total runoff during the hydrological year 2012/2013 (representing 40% of all sources), followed by rainfall (34%) and ice melt (26%). A sensitivity analysis is used to assess the efficiency of the monitoring network and identify the timing and location of field measurements that constrain model uncertainty. The methodology to set up a glacio-hydrological model in high-elevation regions presented in this study can be regarded as a benchmark for modelers in the HKH seeking to evaluate their calibration approach, their experimental setup and thus to reduce the predictive model uncertainty. }, author = {Ragettli, S. and Pellicciotti, Francesca and Immerzeel, W.W. and Miles, E.S. and Petersen, L. and Heynen, M. and Shea, J.M. and Stumm, D. and Joshi, S. and Shrestha, A.}, issn = {0309-1708}, journal = {Advances in Water Resources}, keywords = {Water Science and Technology}, number = {4}, pages = {94--111}, publisher = {Elsevier}, title = {{Unraveling the hydrology of a Himalayan catchment through integration of high resolution in situ data and remote sensing with an advanced simulation model}}, doi = {10.1016/j.advwatres.2015.01.013}, volume = {78}, year = {2015}, } @article{12631, abstract = {Air temperature is one of the most relevant input variables for snow and ice melt calculations. However, local meteorological conditions, complex topography, and logistical concerns in glacierized regions make the measuring and modeling of air temperature a difficult task. In this study, we investigate the spatial distribution of 2 m air temperature over mountain glaciers and propose a modification to an existing model to improve its representation. Spatially distributed meteorological data from Haut Glacier d'Arolla (Switzerland), Place (Canada), and Juncal Norte (Chile) Glaciers are used to examine approximate flow line temperatures during their respective ablation seasons. During warm conditions (off-glacier temperatures well above 0°C), observed air temperatures in the upper reaches of Place Glacier and Haut Glacier d'Arolla decrease down glacier along the approximate flow line. At Juncal Norte and Haut Glacier d'Arolla, an increase in air temperature is observed over the glacier tongue. While the temperature behavior over the upper part can be explained by the cooling effect of the glacier surface, the temperature increase over the glacier tongue may be caused by several processes induced by the surrounding warm atmosphere. In order to capture the latter effect, we add an additional term to the Greuell and Böhm (GB) thermodynamic glacier wind model. For high off-glacier temperatures, the modified GB model reduces root-mean-square error up to 32% and provides a new approach for distributing air temperature over mountain glaciers as a function of off-glacier temperatures and approximate glacier flow lines.}, author = {Ayala, A. and Pellicciotti, Francesca and Shea, J. M.}, issn = {2169-8996}, journal = {Journal of Geophysical Research: Atmospheres}, keywords = {Space and Planetary Science, Earth and Planetary Sciences (miscellaneous), Atmospheric Science, Geophysics}, number = {8}, pages = {3139--3157}, publisher = {American Geophysical Union}, title = {{Modeling 2 m air temperatures over mountain glaciers: Exploring the influence of katabatic cooling and external warming}}, doi = {10.1002/2015jd023137}, volume = {120}, year = {2015}, } @inproceedings{12881, author = {Martius, Georg S and Olbrich, Eckehard}, booktitle = {Proceedings of the 13th European Conference on Artificial Life}, isbn = {9780262330275}, location = {York, United Kingdom}, pages = {78}, publisher = {MIT Press}, title = {{Quantifying self-organizing behavior of autonomous robots}}, doi = {10.7551/978-0-262-33027-5-ch018}, year = {2015}, } @article{1311, abstract = {In this paper, we develop an energy method to study finite speed of propagation and waiting time phenomena for the stochastic porous media equation with linear multiplicative noise in up to three spatial dimensions. Based on a novel iteration technique and on stochastic counterparts of weighted integral estimates used in the deterministic setting, we formulate a sufficient criterion on the growth of initial data which locally guarantees a waiting time phenomenon to occur almost surely. Up to a logarithmic factor, this criterion coincides with the optimal criterion known from the deterministic setting. Our technique can be modified to prove finite speed of propagation as well.}, author = {Julian Fischer and Grün, Günther}, journal = {SIAM Journal on Mathematical Analysis}, number = {1}, pages = {825 -- 854}, publisher = {Society for Industrial and Applied Mathematics }, title = {{Finite speed of propagation and waiting times for the stochastic porous medium equation: A unifying approach}}, doi = {10.1137/140960578}, volume = {47}, year = {2015}, } @article{1313, abstract = {We present an algorithm for the derivation of lower bounds on support propagation for a certain class of nonlinear parabolic equations. We proceed by combining the ideas in some recent papers by the author with the algorithmic construction of entropies due to Jüngel and Matthes, reducing the problem to a quantifier elimination problem. Due to its complexity, the quantifier elimination problem cannot be solved by present exact algorithms. However, by tackling the quantifier elimination problem numerically, in the case of the thin-film equation we are able to improve recent results by the author in the regime of strong slippage n ∈ (1, 2). For certain second-order doubly nonlinear parabolic equations, we are able to extend the known lower bounds on free boundary propagation to the case of irregular oscillatory initial data. Finally, we apply our method to a sixth-order quantum drift-diffusion equation, resulting in an upper bound on the time which it takes for the support to reach every point in the domain.}, author = {Julian Fischer}, journal = {Interfaces and Free Boundaries}, number = {1}, pages = {1 -- 20}, publisher = {European Mathematical Society Publishing House}, title = {{Estimates on front propagation for nonlinear higher-order parabolic equations: An algorithmic approach}}, doi = {10.4171/IFB/331}, volume = {17}, year = {2015}, }