Systematic Strategies and Transfer Effects with Mixed- and Unmixed-List Designs

1970 ◽  
Vol 26 (2) ◽  
pp. 475-484 ◽  
Author(s):  
G. J. Johnson
Keyword(s):  
A Priori ◽  
Mixed List ◽  
Superior Performance ◽  
Transfer Effects ◽  
Training Performance ◽  
Unmixed List ◽  
Transfer Tasks ◽  
Early Training

The present study investigated performance on AB-AB' and AB-ABr paradigm items under mixed list (ML) and unmixed list (UL) conditions. In ML 2 different subsets of second-list items each represented a different type of paradigm, while in UL second-list pairs were homogeneous with respect to type of paradigm. 4 groups of 24 Ss each learned 2 lists of 2-syllable adjectives. First-list items were arranged so as to provide AB-AB' UL, AB-ABr UL and 2 ML conditions. Number of correct responses and trials to criterion measures showed markedly superior performance on AB-ABr items under ML. Early training performance was significantly better for AB-AB' pairs under UL than under ML. The results were viewed as inconsistent with predictions based on a priori assumptions concerning the use of cognitive mediational strategies in transfer tasks. The possible influence of intralist associative factors in ML vs UL transfer effects was discussed.

Pattern of Near Transfer Effects Following Working Memory Training With a Dual N-Back Task

2017 ◽  
Vol 64 (4) ◽  
pp. 240-252 ◽  
Author(s):  
Anna Soveri ◽  
Eric P. A. Karlsson ◽  
Otto Waris ◽  
Petra Grönholm-Nyman ◽  
Matti Laine
Keyword(s):  
Working Memory ◽  
Controlled Trial ◽  
Composite Score ◽  
Memory Training ◽  
Interference Control ◽  
Transfer Effects ◽  
Near Transfer ◽  
Randomized Controlled ◽  
Transfer Tasks ◽  
Back Task

Abstract. In a randomized controlled trial, we investigated the pattern of near transfer effects of working memory (WM) training with an adaptive auditory-visuospatial dual n-back training task in healthy young adults. The results revealed significant task-specific transfer to an untrained single n-back task, and more general near transfer to a WM updating composite score plus a nearly significant effect on a composite score measuring interference control in WM. No transfer effects were seen on Active or Passive WM composites. The results are discussed in the light of cognitive versus strategy-related overlap between training and transfer tasks.

Effects of Non-Anticipation and Anticipation Procedures upon Paired-Associate Learning in Unmixed and Mixed List Designs

1964 ◽  
Vol 15 (3) ◽  
pp. 795-801 ◽  
Author(s):  
James H. Reynolds
Keyword(s):  
Paired Associate ◽  
Mixed List ◽  
Paired Associate Learning ◽  
Associate Learning ◽  
Unmixed List ◽  
Anticipation Method

Two experiments compared verbal PA learning by the standard anticipation technique with learning by a non-anticipation method in which immediate confirmation O- correct responding was eliminated. Most previous investigations have found that learning by the latter procedure is superior to learning by the usual anticipation method. In Exp. I, which employed an unmixed list design, no differences in learning were obtained between the two methods at either of two levels of list difficulty. However, Exp. II, using the same materials in a mixed list design, showed superior learning of items presented by the non-anticipation method regardless of the difficulty of the list. The conflicting results of the two experiments suggest that evidence for superior verbal PA learning by the non-anticipation method may depend, at least in part, upon the list design employed.

L1–2 minimization for P- and S-impedance inversion

2020 ◽  
Vol 8 (2) ◽  
pp. T379-T390
Author(s):  
Wenliang Nie ◽  
Xiaotao Wen ◽  
Jixin Yang ◽  
Jian He ◽  
Kai Lin ◽  
...  
Keyword(s):  
Objective Function ◽  
Reservoir Characterization ◽  
A Priori ◽  
Low Frequency ◽  
Superior Performance ◽  
Minimization Method ◽  
Block Boundary ◽  
Avo Inversion ◽  
Impedance Inversion ◽  
The Difference

Amplitude variation with offset (AVO) inversion has been widely used in reservoir characterization to predict lithology and fluids. However, some existing AVO inversion methods that use [Formula: see text] norm regularization may not obtain the block boundary of subsurface layers because the AVO inversion is a severely ill-posed problem. To obtain sparse and accurate solutions, we have introduced the [Formula: see text] minimization method as an alternative to [Formula: see text] norm regularization. We used [Formula: see text] minimization for simultaneous P- and S-impedance inversion from prestack seismic data. We first derived the forward problem with multiangles and set up the inversion objective function with constraints of a priori low-frequency information obtained from well-log data. Then, we introduced minimization of the difference of [Formula: see text] and [Formula: see text] norms, denoted as [Formula: see text] minimization, to solve this objective function. The nonconvex penalty function of the [Formula: see text] minimization method is decomposed into two convex subproblems via the difference of convex algorithm, and each subproblem is solved by the alternating direction method of multipliers. Compared to [Formula: see text] norm regularization, the results indicate that [Formula: see text] minimization has superior performance over [Formula: see text] norm regularization in promoting blocky/sparse solutions. Tests on synthetic and field data indicate that our method can provide sparser and more accurate P- and S-impedance inversion results. The overall results confirm that our method has great potential in the detection and identification of fluids.

Mixed- versus Unmixed-List Designs in Paired-Associate Learning

1967 ◽  
Vol 20 (3_suppl) ◽  
pp. 1191-1200 ◽  
Author(s):  
Chizuko Izawa
Keyword(s):  
College Students ◽  
Experimental Design ◽  
Paired Associate ◽  
Mixed List ◽  
Paired Associate Learning ◽  
The Other ◽  
Factorial Experimental Design ◽  
Associate Learning ◽  
Unmixed List ◽  
Fluctuation Model

Investigations of a new experimental variable from the arrangements of reinforcements (R) and tests (T) in paired-associate learning were furthered by a 2 × 2 × 2 factorial experimental design: 64 college students learned two lists of 12 pairs, one with unmixed list (Exp. I) and the other with mixed list (Exp. II). Four repetitive experimental sequences in each experiment were RTRT …, RRTRRT. … RTTRTT …, and RRTTRRTT. … No significant differences were found between mixed- and unmixed-list designs for any given statistic examined. The findings indicate that individual pairs in a given condition were learned relatively independently of those in the other conditions within a list. The present results were close replications of the previous study by Izawa (1966a) and support the stimulus fluctuation model.

R-S learning and negative transfer effects with a mixed list.

1963 ◽  
Vol 65 (2) ◽  
pp. 201-205 ◽  
Author(s):  
Donald A. Kausler ◽  
George A. Kanoti
Keyword(s):  
Negative Transfer ◽  
Mixed List ◽  
Transfer Effects

scholarly journals Variational Bayesian Iteration-Based Invariant Kalman Filter for Attitude Estimation on Matrix Lie Groups

Aerospace ◽  
2021 ◽  
Vol 8 (9) ◽  
pp. 246
Author(s):  
Jiaolong Wang ◽  
Zeyang Chen
Keyword(s):  
Kalman Filter ◽  
Lie Groups ◽  
Lie Group ◽  
A Priori ◽  
Attitude Estimation ◽  
Estimation Problem ◽  
Superior Performance ◽  
Process Noise ◽  
Variational Bayesian ◽  
Noise Statistics

Motivated by the rapid progress of aerospace and robotics engineering, the navigation and control systems on matrix Lie groups have been actively studied in recent years. For rigid targets, the attitude estimation problem is a benchmark one with its states defined as rotation matrices on Lie groups. Based on the invariance properties of symmetry groups, the invariant Kalman filter (IKF) has been developed by researchers for matrix Lie group systems; however, the limitation of the IKF is that its estimation performance is prone to be degraded if the given knowledge of the noise statistics is not accurate. For the symmetry Lie group attitude estimation problem, this paper proposes a new variational Bayesian iteration-based adaptive invariant Kalman filter (VBIKF). In the proposed VBIKF, the a priori error covariance is not propagated by the conventional steps but directly calibrated in an iterative manner based on the posterior sequences. The main advantage of the VBIKF is that the statistics parameter of the system process noise is no longer required and so the IKF’s hard dependency on accurate process noise statistics can be reduced significantly. The mathematical foundation for the new VBIKF is presented and its superior performance in adaptability and simplicity is further demonstrated by numerical simulations.

scholarly journals A Temporal Directed Graph Convolution Network for Traffic Forecasting Using Taxi Trajectory Data

2021 ◽  
Vol 10 (9) ◽  
pp. 624
Author(s):  
Kaiqi Chen ◽  
Min Deng ◽  
Yan Shi
Keyword(s):  
Traffic Flow ◽  
Directed Graph ◽  
Intelligent Transportation Systems ◽  
Traffic Management ◽  
A Priori ◽  
Transportation Systems ◽  
Superior Performance ◽  
Trajectory Data ◽  
Temporal Dependence ◽  
Traffic Forecasting

Traffic forecasting plays a vital role in intelligent transportation systems and is of great significance for traffic management. The main issue of traffic forecasting is how to model spatial and temporal dependence. Current state-of-the-art methods tend to apply deep learning models; these methods are unexplainable and ignore the a priori characteristics of traffic flow. To address these issues, a temporal directed graph convolution network (T-DGCN) is proposed. A directed graph is first constructed to model the movement characteristics of vehicles, and based on this, a directed graph convolution operator is used to capture spatial dependence. For temporal dependence, we couple a keyframe sequence and transformer to learn the tendencies and periodicities of traffic flow. Using a real-world dataset, we confirm the superior performance of the T-DGCN through comparative experiments. Moreover, a detailed discussion is presented to provide the path of reasoning from the data to the model design to the conclusions.

scholarly journals Hedging the Drift: Learning to Optimize Under Nonstationarity

2021 ◽  
Author(s):  
Wang Chi Cheung ◽  
David Simchi-Levi ◽  
Ruihao Zhu
Keyword(s):  
Dynamic Pricing ◽  
A Priori ◽  
Network Routing ◽  
Data Driven ◽  
Superior Performance ◽  
Changing Environments ◽  
Epidemic Period ◽  
Optimal Dynamic ◽  
Regret Bounds ◽  
Upper Confidence Bound

We introduce data-driven decision-making algorithms that achieve state-of-the-art dynamic regret bounds for a collection of nonstationary stochastic bandit settings. These settings capture applications such as advertisement allocation, dynamic pricing, and traffic network routing in changing environments. We show how the difficulty posed by the (unknown a priori and possibly adversarial) nonstationarity can be overcome by an unconventional marriage between stochastic and adversarial bandit learning algorithms. Beginning with the linear bandit setting, we design and analyze a sliding window-upper confidence bound algorithm that achieves the optimal dynamic regret bound when the underlying variation budget is known. This budget quantifies the total amount of temporal variation of the latent environments. Boosted by the novel bandit-over-bandit framework that adapts to the latent changes, our algorithm can further enjoy nearly optimal dynamic regret bounds in a (surprisingly) parameter-free manner. We extend our results to other related bandit problems, namely the multiarmed bandit, generalized linear bandit, and combinatorial semibandit settings, which model a variety of operations research applications. In addition to the classical exploration-exploitation trade-off, our algorithms leverage the power of the “forgetting principle” in the learning processes, which is vital in changing environments. Extensive numerical experiments with synthetic datasets and a dataset of an online auto-loan company during the severe acute respiratory syndrome (SARS) epidemic period demonstrate that our proposed algorithms achieve superior performance compared with existing algorithms. This paper was accepted by George J. Shanthikumar for the Management Science Special Issue on Data-Driven Prescriptive Analytics.

Sign in / Sign up

Export Citation Format

Share Document