Training Deep Architectures Without End-to-End Backpropagation: A Survey on the Provably Optimal Methods

This tutorial paper surveys provably optimal alternatives to end-to-end backpropagation (E2EBP) — the de facto standard for training deep architectures. Modular training refers to strictly local training without both the forward and the backward pass, i.e., dividing a deep architecture into several nonoverlapping modules and training them separately without any end-to-end operation. Between the fully global E2EBP and the strictly local modular training, there are weakly modular hybrids performing training without the backward pass only. Read More

IEEE Computational Intelligence Magazine, November 2022

On Mixed-Initiative Content Creation for Video Games

In this article, we present a survey of mixed-initiative methods for the creation of content for video games. We also propose a definition of what mixed initiative implies, as the term lacks a clear specification. The survey includes works not directly aimed at video games but which create content that can potentially be used in games, such as art programs utilizing mixed initiative. Furthermore, we highlight research areas that overlap wholly or partly with mixed initiative, such as casual creators, explainable artificial intelligence, or interactive evolutionary computation. Read More

IEEE Transactions on Games, December 2022

Supervised Learning in Neural Networks: Feedback-Network-Free Implementation and Biological Plausibility

The well-known backpropagation learning algorithm is probably the most popular learning algorithm in artificial neural networks. It has been widely used in various applications of deep learning. The backpropagation algorithm requires a separate feedback network to back propagate errors. This feedback network must have the same topology and connection strengths (weights) as the feed-forward network. In this article, we propose a new learning algorithm that is mathematically equivalent to the backpropagation algorithm but does not require a feedback network.  Read more

IEEE Transactions on Neural Networks and Learning Systems, December 2022

FSNet: Dual Interpretable Graph Convolutional Network for Alzheimer's Disease Analysis

Graph Convolutional Networks (GCNs) are widely used in medical images diagnostic research, because they can automatically learn powerful and robust feature representations. However, their performance might be significantly deteriorated by trivial or corrupted medical features and samples. Moreover, existing methods cannot simultaneously interpret the significant features and samples. To overcome these limitations, in this paper, we propose a novel dual interpretable graph convolutional network, namely FSNet, to simultaneously select significant features and samples, so as to boost model performance for medical diagnosis and interpretation. Read More

IEEE Transactions on Emerging Topics in Computational Intelligence, July 2022

Anytime Performance Assessment in Blackbox Optimization Benchmarking

We present concepts and recipes for the anytime performance assessment when benchmarking optimization algorithms in a blackbox scenario. We consider runtime—oftentimes measured in the number of blackbox evaluations needed to reach a target quality—to be a universally measurable cost for solving a problem. Starting from the graph that depicts the solution quality versus runtime, we argue that runtime is the only performance measure with a generic, meaningful, and quantitative interpretation. Hence, our assessment is solely based on runtime measurements. Read More

IEEE Transactions on Evolutionary Computation, December 2022

The Challenges and Opportunities of Human-Centered AI for Trustworthy Robots and Autonomous Systems

The trustworthiness of robots and autonomous systems (RASs) has taken a prominent position on the way toward full autonomy. This work is the first to systematically explore the key facets of human-centered AI (HAI) for trustworthy RAS (TRAS). We identified five key properties of a TRAS, i.e., RAS must be: 1) safe in any uncertain and dynamic environment; 2) secure, i.e., protect itself from cyber threats; 3) healthy and fault tolerant; 4) trusted and easy to use to enable effective human–machine interaction (HMI); and 5) compliant with the law and ethical expectations. Read More

IEEE Transactions on Cognitive and Developmental Systems, December 2022

Large-Scale Fuzzy Least Squares Twin SVMs for Class Imbalance Learning

Twin support vector machines (TSVMs) have been successfully employed for binary classification problems. With the advent of machine learning algorithms, data have proliferated and there is a need to handle or process large-scale data. TSVMs are not successful in handling large-scale data due to the following: 1) the optimization problem solved in the TSVM needs to calculate large matrix inverses, which makes it an ineffective choice for large-scale problems; 2) the empirical risk minimization principle is employed in the TSVM and, hence, may suffer due to overfitting; and 3) the Wolfe dual of TSVM formulation involves positive-semidefinite matrices, and hence, singularity issues need to be resolved manually. Read More

IEEE Transactions on Fuzzy Systems, November 2022

An Improved Explainable Point Cloud Classifier (XPCC)

Classification of objects from 3-D point clouds has become an increasingly relevant task across many computer-vision applications. However, few studies have investigated explainable methods. In this article, a new prototype-based and explainable classification method called eXplainable point cloud classifier (XPCC) is proposed. The XPCC method offers several advantages over previous explainable and nonexplainable methods. Read More

IEEE Transactions on Artificial Intelligence, February 2023

• IEEE TNNLS Special Issue on Graphic Learning (1 March 2023)
  

• IEEE TEVC Special Issue on Evolutionary Neural Architecture Search (28 March 2023)

• IEEE TNNLS Special Issue on Learning Theories and Methods with Application to Digitized Process Manufacturing (31 March 2023)

• IEEE TNNLS Special Issue on Deep Learning: When and How? (23 April 2023 )

• IEEE TNNLS Special issue on Trustworthy Federated Learning (1 June 2023)

• IEEE TAI Special Issue on AutoML for Non-Stationary Data (2 June 2023)

FUZZ-IEEE 2023 Call for Papers

The Annual IEEE International Conference on Fuzzy Systems, well known as FUZZ-IEEE, is one of the premier international conferences in the field of fuzzy sets and systems. FUZZ-IEEE 2023 will be held in the Songdo International City which is a Free Economic Zone located approximately 15 minutes away by car from the Incheon International Airport and an hour from Seoul, making it a pleasant and convenient venue for business as well as for touring the Seoul metropolitan area. It is rising as one of the hottest places in Korea in terms of tourism, culture, leisure, MICE (Meeting, Incentive trip, Convention, Exhibition & Event), and shopping. This complex cultural space provides sharing of happiness to all visitors. Access all submission details here.

IEEE CAI 2023 Call for Papers

The IEEE Conference on Artificial Intelligence (IEEE CAI) will take place in Santa Clara, CA, USA on 5-6 June 2023. This new conference series emphasizes AI applications in specific verticals that impact industrial technology and innovation. Key verticals include Energy, Healthcare and Life Science, Transportation and Aerospace, Earth Systems Decision Support, Industrial AI, and Ethical and Societal Implications of AI. IEEE CAI seeks original, high-quality poster papers describing the research and results that contribute to advancements in AI applications. Access all submission details here.