-
A Feature-based Generalizable Prediction Model for Both Perceptual and Abstract Reasoning
Authors:
Quan Do,
Thomas M. Morin,
Chantal E. Stern,
Michael E. Hasselmo
Abstract:
A hallmark of human intelligence is the ability to infer abstract rules from limited experience and apply these rules to unfamiliar situations. This capacity is widely studied in the visual domain using the Raven's Progressive Matrices. Recent advances in deep learning have led to multiple artificial neural network models matching or even surpassing human performance. However, while humans can ide…
▽ More
A hallmark of human intelligence is the ability to infer abstract rules from limited experience and apply these rules to unfamiliar situations. This capacity is widely studied in the visual domain using the Raven's Progressive Matrices. Recent advances in deep learning have led to multiple artificial neural network models matching or even surpassing human performance. However, while humans can identify and express the rule underlying these tasks with little to no exposure, contemporary neural networks often rely on massive pattern-based training and cannot express or extrapolate the rule inferred from the task. Furthermore, most Raven's Progressive Matrices or Raven-like tasks used for neural network training used symbolic representations, whereas humans can flexibly switch between symbolic and continuous perceptual representations. In this work, we present an algorithmic approach to rule detection and application using feature detection, affine transformation estimation and search. We applied our model to a simplified Raven's Progressive Matrices task, previously designed for behavioral testing and neuroimaging in humans. The model exhibited one-shot learning and achieved near human-level performance in the symbolic reasoning condition of the simplified task. Furthermore, the model can express the relationships discovered and generate multi-step predictions in accordance with the underlying rule. Finally, the model can reason using continuous patterns. We discuss our results and their relevance to studying abstract reasoning in humans, as well as their implications for improving intelligent machines.
△ Less
Submitted 8 March, 2024;
originally announced March 2024.
-
Frugal Splitting Operators: Representation, Minimal Lifting, and Convergence
Authors:
Martin Morin,
Sebastian Banert,
Pontus Giselsson
Abstract:
We investigate frugal splitting operators for finite sum monotone inclusion problems. These operators utilize exactly one direct or resolvent evaluation of each operator of the sum, and the splitting operator's output is dictated by linear combinations of these evaluations' inputs and outputs. To facilitate analysis, we introduce a novel representation of frugal splitting operators via a generaliz…
▽ More
We investigate frugal splitting operators for finite sum monotone inclusion problems. These operators utilize exactly one direct or resolvent evaluation of each operator of the sum, and the splitting operator's output is dictated by linear combinations of these evaluations' inputs and outputs. To facilitate analysis, we introduce a novel representation of frugal splitting operators via a generalized primal-dual resolvent. The representation is characterized by an index and four matrices, and we provide conditions on these that ensure equivalence between the classes of frugal splitting operators and generalized primal-dual resolvents. Our representation paves the way for new results regarding lifting numbers and the development of a unified convergence analysis for frugal splitting operator methods, contingent on the directly evaluated operators being cocoercive. The minimal lifting number is $n-1-f$ where $n$ is the number of monotone operators and $f$ is the number of direct evaluations in the splitting. Notably, this lifting number is achievable only if the first and last operator evaluations are resolvent evaluations. These results generalize the minimal lifting results by Ryu and Malitsky--Tam that consider frugal resolvent splittings. Building on our representation, we delineate a constructive method to design frugal splitting operators, exemplified in the design of a novel, convergent, and parallelizable frugal splitting operator with minimal lifting.
△ Less
Submitted 29 September, 2023; v1 submitted 22 June, 2022;
originally announced June 2022.
-
Weak ferromagnetism linked to the high-temperature spiral phase of YBaCuFeO5
Authors:
J. Lyu,
M. Morin,
T. Shang,
M. T. Fernandez-Diaz,
M. Medarde
Abstract:
The layered perovskite YBaCuFeO5 is a rare example of cycloidal spiral magnet whose ordering temperature Tspiral can be tuned far beyond room temperature by adjusting the degree of Cu2+/Fe3+ chemical disorder in the structure. This unusual property qualifies this material as one of the most promising spin-driven multiferroic candidates. However, very little is known about the response of the spira…
▽ More
The layered perovskite YBaCuFeO5 is a rare example of cycloidal spiral magnet whose ordering temperature Tspiral can be tuned far beyond room temperature by adjusting the degree of Cu2+/Fe3+ chemical disorder in the structure. This unusual property qualifies this material as one of the most promising spin-driven multiferroic candidates. However, very little is known about the response of the spiral to magnetic fields, crucial for magnetoelectric cross-control applications. Using bulk magnetization and neutron powder diffraction measurements under magnetic fields up to 9 T we report here the first temperature-magnetic field phase diagram of this material. Besides revealing a strong stability of the spiral state, our data uncover the presence of weak ferromagnetism coexisting with the spiral modulation. Since ferromagnets can be easily manipulated with magnetic fields, this observation opens new perspectives for the control of the spiral orientation, directly linked to the polarization direction, as well as for a possible future use of this material in technological applications.
△ Less
Submitted 22 March, 2022;
originally announced March 2022.
-
Nonlinear Forward-Backward Splitting with Momentum Correction
Authors:
Martin Morin,
Sebastian Banert,
Pontus Giselsson
Abstract:
The nonlinear, or warped, resolvent recently explored by Giselsson and Bùi-Combettes has been used to model a large set of existing and new monotone inclusion algorithms. To establish convergent algorithms based on these resolvents, corrective projection steps are utilized in both works. We present a different way of ensuring convergence by means of a nonlinear momentum term, which in many cases l…
▽ More
The nonlinear, or warped, resolvent recently explored by Giselsson and Bùi-Combettes has been used to model a large set of existing and new monotone inclusion algorithms. To establish convergent algorithms based on these resolvents, corrective projection steps are utilized in both works. We present a different way of ensuring convergence by means of a nonlinear momentum term, which in many cases leads to cheaper per-iteration cost. The expressiveness of our method is demonstrated by deriving a wide range of special cases. These cases cover and expand on the forward-reflected-backward method of Malitsky-Tam, the primal-dual methods of Vũ-Condat and Chambolle-Pock, and the forward-reflected-Douglas-Rachford method of Ryu-Vũ. A new primal-dual method that uses an extra resolvent step is also presented as well as a general approach for adding momentum to any special case of our nonlinear forward-backward method, in particular all the algorithms listed above.
△ Less
Submitted 29 September, 2023; v1 submitted 1 December, 2021;
originally announced December 2021.
-
Sampling and Update Frequencies in Proximal Variance-Reduced Stochastic Gradient Methods
Authors:
Martin Morin,
Pontus Giselsson
Abstract:
Variance-reduced stochastic gradient methods have gained popularity in recent times. Several variants exist with different strategies for the storing and sampling of gradients and this work concerns the interactions between these two aspects. We present a general proximal variance-reduced gradient method and analyze it under strong convexity assumptions. Special cases of the algorithm include SAGA…
▽ More
Variance-reduced stochastic gradient methods have gained popularity in recent times. Several variants exist with different strategies for the storing and sampling of gradients and this work concerns the interactions between these two aspects. We present a general proximal variance-reduced gradient method and analyze it under strong convexity assumptions. Special cases of the algorithm include SAGA, L-SVRG and their proximal variants. Our analysis sheds light on epoch-length selection and the need to balance the convergence of the iterates with how often gradients are stored. The analysis improves on other convergence rates found in the literature and produces a new and faster converging sampling strategy for SAGA. Problem instances for which the predicted rates are the same as the practical rates are presented together with problems based on real world data.
△ Less
Submitted 18 October, 2022; v1 submitted 13 February, 2020;
originally announced February 2020.
-
Non-Determinism in TensorFlow ResNets
Authors:
Miguel Morin,
Matthew Willetts
Abstract:
We show that the stochasticity in training ResNets for image classification on GPUs in TensorFlow is dominated by the non-determinism from GPUs, rather than by the initialisation of the weights and biases of the network or by the sequence of minibatches given. The standard deviation of test set accuracy is 0.02 with fixed seeds, compared to 0.027 with different seeds---nearly 74\% of the standard…
▽ More
We show that the stochasticity in training ResNets for image classification on GPUs in TensorFlow is dominated by the non-determinism from GPUs, rather than by the initialisation of the weights and biases of the network or by the sequence of minibatches given. The standard deviation of test set accuracy is 0.02 with fixed seeds, compared to 0.027 with different seeds---nearly 74\% of the standard deviation of a ResNet model is non-deterministic. For test set loss the ratio of standard deviations is more than 80\%. These results call for more robust evaluation strategies of deep learning models, as a significant amount of the variation in results across runs can arise simply from GPU randomness.
△ Less
Submitted 30 January, 2020;
originally announced January 2020.
-
Cocoercivity, Smoothness and Bias in Variance-Reduced Stochastic Gradient Methods
Authors:
Martin Morin,
Pontus Giselsson
Abstract:
With the purpose of examining biased updates in variance-reduced stochastic gradient methods, we introduce SVAG, a SAG/SAGA-like method with adjustable bias. SVAG is analyzed in a cocoercive root-finding setting, a setting which yields the same results as in the usual smooth convex optimization setting for the ordinary proximal-gradient method. We show that the same is not true for SVAG when biase…
▽ More
With the purpose of examining biased updates in variance-reduced stochastic gradient methods, we introduce SVAG, a SAG/SAGA-like method with adjustable bias. SVAG is analyzed in a cocoercive root-finding setting, a setting which yields the same results as in the usual smooth convex optimization setting for the ordinary proximal-gradient method. We show that the same is not true for SVAG when biased updates are used. The step-size requirements for when the operators are gradients are significantly less restrictive compared to when they are not. This highlights the need to not rely solely on cocoercivity when analyzing variance-reduced methods meant for optimization. Our analysis either match or improve on previously known convergence conditions for SAG and SAGA. However, in the biased cases they still do not correspond well with practical experiences and we therefore examine the effect of bias numerically on a set of classification problems. The choice of bias seem to primarily affect the early stages of convergence and in most cases the differences vanish in the later stages of convergence. However, the effect of the bias choice is still significant in a couple of cases.
△ Less
Submitted 1 February, 2022; v1 submitted 21 March, 2019;
originally announced March 2019.
-
Etude d'un dispositif d'aide a l'intention d'{é}l{è}ves en difficult{é} dans la r{é}solution d'une situation-probl{è}me math{é}matique
Authors:
Marie-Pier Morin,
Laurent Theis,
Teresa Assude,
Karine Millon-Fauré,
Jeannette Tambone,
Jeanne Koudogbo,
Valérie Hamel,
Patricia Marchand
Abstract:
In this paper, we analyze an aid session tested by an elementary school teachers. This aid session has been set up by a teacher for some students with difficulties after the work in the whole class. We first show how this aid session can help the pupils, especially through its chrono-topo-meso-genetic functions. Then we analyze a second situation in which the aid session is implemented, looking fo…
▽ More
In this paper, we analyze an aid session tested by an elementary school teachers. This aid session has been set up by a teacher for some students with difficulties after the work in the whole class. We first show how this aid session can help the pupils, especially through its chrono-topo-meso-genetic functions. Then we analyze a second situation in which the aid session is implemented, looking for clues of these functions in some episodes involving students who participated in the aid session.
△ Less
Submitted 18 March, 2019;
originally announced March 2019.
-
Design of magnetic spirals in layered perovskites: extending the stability range far beyond room temperature
Authors:
Tian Shang,
Emmanuel Canévet,
Mickaël Morin,
Denis Sheptyakov,
María Teresa Fernández-Díaz,
Ekaterina Pomjakushina,
Marisa Medarde
Abstract:
In insulating materials with ordered magnetic spiral phases, ferroelectricity can emerge due to the breaking of inversion symmetry. This property is of both fundamental and practical interest, in particular with a view to exploiting it in low-power electronic devices. Advances towards technological applications have been hindered, however, by the relatively low ordering temperatures…
▽ More
In insulating materials with ordered magnetic spiral phases, ferroelectricity can emerge due to the breaking of inversion symmetry. This property is of both fundamental and practical interest, in particular with a view to exploiting it in low-power electronic devices. Advances towards technological applications have been hindered, however, by the relatively low ordering temperatures $T_\mathrm{spiral}$ of most magnetic spiral phases, which rarely exceed 100 K. We have recently established that the ordering temperature of a magnetic spiral can be increased up to 310 K by the introduction of chemical disorder. Here we explore the design space opened up by this novel mechanism by combining it with a targeted lattice control of some magnetic interactions. In Cu-Fe layered perovskites we obtain $T_\mathrm{spiral}$ values close to 400 K, comfortably far from room temperature and almost 100 K higher than using chemical disorder alone. Moreover, we reveal a linear, universal relationship between the spiral's wave vector and the onset temperature of the spiral phase. This linear law ends at a paramagnetic-collinear-spiral triple point, which defines the highest spiral ordering temperature that can be achieved in this class of materials. Based on these findings, we propose a general set of rules for designing magnetic spirals in layered perovskites using external pressure, chemical substitutions and/or epitaxial strain, which should guide future efforts to engineer magnetic spiral phases with ordering temperatures suitable for technological applications.
△ Less
Submitted 27 September, 2018;
originally announced September 2018.
-
An iterative closest point method for measuring the level of similarity of 3d log scans in wood industry
Authors:
Cyrine Selma,
Hind Haouzi,
Philippe Thomas,
Jonathan Gaudreault,
Michael Morin
Abstract:
In the Canadian's lumber industry, simulators are used to predict the lumbers resulting from the sawing of a log at a given sawmill. Giving a log or several logs' 3D scans as input, simulators perform a real-time job to predict the lumbers. These simulators, however, tend to be slow at processing large volume of wood. We thus explore an alternative approximation techniques based on the Iterative C…
▽ More
In the Canadian's lumber industry, simulators are used to predict the lumbers resulting from the sawing of a log at a given sawmill. Giving a log or several logs' 3D scans as input, simulators perform a real-time job to predict the lumbers. These simulators, however, tend to be slow at processing large volume of wood. We thus explore an alternative approximation techniques based on the Iterative Closest Point (ICP) algorithm to identify the already processed log to which an unseen log resembles the most. The main benefit of the ICP approach is that it can easily handle 3D scans with a variable number of points. We compare this ICP-based nearest neighbor predictor, to predictors built using machine learning algorithms such as the K-nearest-neighbor (kNN) and Random Forest (RF). The implemented ICP-based predictor enabled us to identify key points in using the 3D scans directly for distance calculation. The long-term goal of this ongoing research is to integrated ICP distance calculations and machine learning.
△ Less
Submitted 23 October, 2017;
originally announced October 2017.
-
Tuning magnetic spirals beyond room temperature with chemical disorder
Authors:
Mickaël Morin,
Emmanuel Canévet,
Adrien Raynaud,
Marek Bartkowiak,
Denis Sheptyakov,
Voraksmy Ban,
Michel Kenzelmann,
Ekaterina Pomjakushina,
Kazimierz Conder,
Marisa Medarde
Abstract:
In the past years, magnetism-driven ferroelectricity and gigantic magnetoelectric effects have been reported for a number of frustrated magnets with spiral magnetic orders. Such materials are of high current interest due to their potential for spintronics and low-power magnetoelectric devices. However, their low magnetic order temperatures (typically <100K) greatly restrict their fields of applica…
▽ More
In the past years, magnetism-driven ferroelectricity and gigantic magnetoelectric effects have been reported for a number of frustrated magnets with spiral magnetic orders. Such materials are of high current interest due to their potential for spintronics and low-power magnetoelectric devices. However, their low magnetic order temperatures (typically <100K) greatly restrict their fields of application. Here we demonstrate that the stability domain of the spiral phase in the perovskite YBaCuFeO5 can be enlarged by more than 150K through a controlled manipulation of the Fe/Cu chemical disorder. Moreover we show that this novel mechanism can stabilize the magnetic spiral state of YBaCuFeO5 above the symbolic value of 25°C at zero magnetic field. Our findings demonstrate that the properties of a magnetic spiral, including its wavelength and stability range, can be engineered through the control of chemical disorder, offering a great potential for the design of materials with magnetoelectric properties beyond room temperature.
△ Less
Submitted 13 January, 2017; v1 submitted 11 August, 2016;
originally announced August 2016.
-
Incommensurate magnetic structure, Fe/Cu chemical disorder and magnetic interactions in the high-temperature multiferroic YBaCuFeO5
Authors:
M. Morin,
A. Scaramucci,
M. Bartkowiak,
E. Pomjakushina,
G. Deng,
D. Sheptyakov,
L. Keller,
J. Rodriguez-Carvajal,
N. A. Spaldin,
M. Kenzelmann,
K. Conder,
M. Medarde
Abstract:
Motivated by the recent observations of incommensurate magnetic order and electric polarization in YBaCuFeO5 up to temperatures TN2 as high as 230K [1,2] we report here for the first time a model for the incommensurate magnetic structure of this material that we complement with ab-initio calculations of the magnetic exchange parameters. Using neutron powder diffraction we show that the appearance…
▽ More
Motivated by the recent observations of incommensurate magnetic order and electric polarization in YBaCuFeO5 up to temperatures TN2 as high as 230K [1,2] we report here for the first time a model for the incommensurate magnetic structure of this material that we complement with ab-initio calculations of the magnetic exchange parameters. Using neutron powder diffraction we show that the appearance of polarization below TN2 is accompanied by the replacement of the high temperature collinear magnetic order by a circular inclined spiral with propagation vector ki=(1/2, 1/2, 1/2+-q). Moreover, we find that the polarization approximately scales with the modulus of the magnetic modulation vector q down to the lowest temperature investigated (T=3K). Further, we observe occupational Fe/Cu disorder in the FeO5-CuO5 bipyramids, although a preferential occupation of such units by Fe-Cu pairs is supported by the observed magnetic order and by density functional calculations. We calculate exchange coupling constants for different Fe/Cu distributions and show that, for those containing Fe-Cu dimers, the resulting magnetic order is compatible with the experimentally observed collinear magnetic structure (kc=(1/2, 1/2, 1/2), TN2 > T > TN1 = 440K). Based on these results we discuss possible origins for the incommensurate modulation and its coupling with ferroelectricity.
△ Less
Submitted 8 January, 2015;
originally announced January 2015.
-
Differences of Augmented Staircase Skew Schur Functions
Authors:
Matthew Morin
Abstract:
We define a fat staircase to be a Ferrers diagram corresponding to a partition of the form $(n^{α_n}, {n-1}^{α_{n-1}},..., 1^{α_1})$, where $α= (α_1,...,α_n)$ is a composition, or the $180^\circ$ rotation of such a diagram. We look at collections of skew diagrams consisting of a fixed fat staircase augmented with all hooks of a given size. Among these diagrams we determine precisely which pairs gi…
▽ More
We define a fat staircase to be a Ferrers diagram corresponding to a partition of the form $(n^{α_n}, {n-1}^{α_{n-1}},..., 1^{α_1})$, where $α= (α_1,...,α_n)$ is a composition, or the $180^\circ$ rotation of such a diagram. We look at collections of skew diagrams consisting of a fixed fat staircase augmented with all hooks of a given size. Among these diagrams we determine precisely which pairs give a Schur-positive difference. We extend this classification to collections of fat staircases augmented with hook-complements.
△ Less
Submitted 29 March, 2010;
originally announced March 2010.
-
Skew Schur Functions of Sums of Fat Staircases
Authors:
Matthew Morin
Abstract:
We define a fat staircase to be a Ferrers diagram corresponding to a partition of the form $(n^{α_n}, {n-1}^{α_{n-1}},..., 1^{α_1})$, where $α= (α_1,...,α_n)$ is a composition, or the $180^\circ$ rotation of such a diagram. If a diagram's skew Schur function is a linear combination of Schur functions of fat staircases, we call the diagram a sum of fat staircases. We prove a Schur-positivity result…
▽ More
We define a fat staircase to be a Ferrers diagram corresponding to a partition of the form $(n^{α_n}, {n-1}^{α_{n-1}},..., 1^{α_1})$, where $α= (α_1,...,α_n)$ is a composition, or the $180^\circ$ rotation of such a diagram. If a diagram's skew Schur function is a linear combination of Schur functions of fat staircases, we call the diagram a sum of fat staircases. We prove a Schur-positivity result that is obtained each time we augment a sum of fat staircases with a skew diagram. We also determine conditions on which diagrams can be sums of fat staircases, including necessary and sufficient conditions in the special case when the diagram is a fat staircase skew a single row or column.
△ Less
Submitted 8 March, 2010;
originally announced March 2010.
-
Differences of Skew Schur Functions of Staircases with Transposed Foundations
Authors:
Matthew Morin
Abstract:
We consider the skew diagram $Δ_n$, which is the $180^\circ$ rotation of the staircase diagram $δ_n = (n,n-1,n-2,...,2,1)$. We create a staircase with bad foundation by augmenting $Δ_n$ with another skew diagram, which we call the \textit{foundation}. We consider pairs of staircases with bad foundation whose foundations are transposes of one another. Among these pairs, we show that the difference…
▽ More
We consider the skew diagram $Δ_n$, which is the $180^\circ$ rotation of the staircase diagram $δ_n = (n,n-1,n-2,...,2,1)$. We create a staircase with bad foundation by augmenting $Δ_n$ with another skew diagram, which we call the \textit{foundation}. We consider pairs of staircases with bad foundation whose foundations are transposes of one another. Among these pairs, we show that the difference of the corresponding skew Schur functions is Schur-positive in the case when one of the foundations consists of either a one or two row diagram, or a hook diagram.
△ Less
Submitted 8 March, 2010;
originally announced March 2010.
-
On distinguishing trees by their chromatic symmetric functions
Authors:
Jeremy L. Martin,
Matthew Morin,
Jennifer D. Wagner
Abstract:
Let $T$ be an unrooted tree. The \emph{chromatic symmetric function} $X_T$, introduced by Stanley, is a sum of monomial symmetric functions corresponding to proper colorings of $T$. The \emph{subtree polynomial} $S_T$, first considered under a different name by Chaudhary and Gordon, is the bivariate generating function for subtrees of $T$ by their numbers of edges and leaves. We prove that…
▽ More
Let $T$ be an unrooted tree. The \emph{chromatic symmetric function} $X_T$, introduced by Stanley, is a sum of monomial symmetric functions corresponding to proper colorings of $T$. The \emph{subtree polynomial} $S_T$, first considered under a different name by Chaudhary and Gordon, is the bivariate generating function for subtrees of $T$ by their numbers of edges and leaves. We prove that $S_T = <Φ,X_T>$, where $<\cdot,\cdot>$ is the Hall inner product on symmetric functions and $Φ$ is a certain symmetric function that does not depend on $T$. Thus the chromatic symmetric function is a stronger isomorphism invariant than the subtree polynomial. As a corollary, the path and degree sequences of a tree can be obtained from its chromatic symmetric function. As another application, we exhibit two infinite families of trees (\emph{spiders} and some \emph{caterpillars}), and one family of unicyclic graphs (\emph{squids}) whose members are determined completely by their chromatic symmetric functions.
△ Less
Submitted 8 June, 2007; v1 submitted 12 September, 2006;
originally announced September 2006.
-
Temperature and magnetic field dependences of the elastic constants of Ni-Mn-Al magnetic Heusler alloys
Authors:
Xavier Moya,
Lluis Manosa,
Antoni Planes,
Thorsten Krenke,
Mehmet Acet,
Michel Morin,
J. L. Zarestky,
T. A. Lograsso
Abstract:
We report on measurements of the adiabatic second order elastic constants of the off-stoichiometric Ni$_{54}$Mn$_{23}$Al$_{23}$ single crystalline Heusler alloy. The variation in the temperature dependence of the elastic constants has been investigated across the magnetic transition and over a broad temperature range. Anomalies in the temperature behaviour of the elastic constants have been foun…
▽ More
We report on measurements of the adiabatic second order elastic constants of the off-stoichiometric Ni$_{54}$Mn$_{23}$Al$_{23}$ single crystalline Heusler alloy. The variation in the temperature dependence of the elastic constants has been investigated across the magnetic transition and over a broad temperature range. Anomalies in the temperature behaviour of the elastic constants have been found in the vicinity of the magnetic phase transition. Measurements under applied magnetic field, both isothermal and variable temperature, show that the value of the elastic constants depends on magnetic order, thus giving evidence for magnetoelastic coupling in this alloy system.
△ Less
Submitted 2 June, 2006;
originally announced June 2006.
-
Disorder induced critical phenomena in magnetically glassy Cu-Al-Mn alloys
Authors:
Jordi Marcos,
Eduard Vives,
Lluis Manosa,
Mehmet Acet,
Eyup Duman,
Michel Morin,
Vaclav Novak,
Antoni Planes
Abstract:
Measurements of magnetic hysteresis loops in Cu-Al-Mn alloys of different Mn content at low temperatures are presented. The loops are smooth and continuous above a certain temperature, but exhibit a magnetization discontinuity below that temperature. Scaling analysis suggest that this system displays a disorder induced phase transition line. Measurements allow to determine the critical exponents…
▽ More
Measurements of magnetic hysteresis loops in Cu-Al-Mn alloys of different Mn content at low temperatures are presented. The loops are smooth and continuous above a certain temperature, but exhibit a magnetization discontinuity below that temperature. Scaling analysis suggest that this system displays a disorder induced phase transition line. Measurements allow to determine the critical exponents $β=0.03\pm 0.01$ and $βδ= 0.4 \pm 0.1$ in agreement with those reported recently [Berger et al., Phys. Rev. Lett. {\bf 85}, 4176 (2000)]
△ Less
Submitted 13 September, 2002;
originally announced September 2002.
