Kernel functions are vital ingredients of several machine learning algorithms, but often incur significant memory and computational costs. We introduce an approach to kernel approximation in machine learning algorithms suitable for mixed-signal Analog In-Memory Computing (AIMC) architectures. Analog In-Memory Kernel Approximation addresses the performance bottlenecks of conventional kernel-based methods by executing most operations in approximate kernel methods directly in memory. The IBM HERMES Project Chip, a state-of-the-art phase-change memory based AIMC chip, is utilized for the hardware demonstration of kernel approximation. Experimental results show that our method maintains high accuracy, with less than a 1% drop in kernel-based ridge classification benchmarks and within 1% accuracy on the Long Range Arena benchmark for kernelized attention in Transformer neural networks. Compared to traditional digital accelerators, our approach is estimated to deliver superior energy efficiency and lower power consumption. These findings highlight the potential of heterogeneous AIMC architectures to enhance the efficiency and scalability of machine learning applications.

Graph neural networks (GNNs) encounter significant computational challenges when handling large-scale graphs, which severely restricts their efficacy across diverse applications. To address this limitation, graph condensation has emerged as a promising technique, which constructs a small synthetic graph for efficiently training GNNs while retaining performance. However, due to the topology structure among nodes, graph condensation is limited to condensing only the observed training nodes and their corresponding structure, thus lacking the ability to effectively handle the unseen data. Consequently, the original large graph is still required in the inference stage to perform message passing to inductive nodes, resulting in substantial computational demands. To overcome this issue, we propose mapping-aware graph condensation (MCond), explicitly learning the one-to-many node mapping from original nodes to synthetic nodes to seamlessly integrate new nodes into the synthetic graph for inductive representation learning. This enables direct information propagation on the synthetic graph, which is much more efficient than on the original large graph. Specifically, MCond employs an alternating optimization scheme with innovative loss terms from transductive and inductive perspectives, facilitating the mutual promotion between graph condensation and node mapping learning. Extensive experiments demonstrate the efficacy of our approach in inductive inference. On the Reddit dataset, MCond achieves up to 121.5x inference speedup and 55.9x reduction in storage requirements compared with counterparts based on the original graph.

Named Entity Recognition (NER) frequently suffers from the problem of insufficient labeled data, particularly in fine-grained NER scenarios. Although $K$-shot learning techniques can be applied, their performance tends to saturate when the number of annotations exceeds several tens of labels. To overcome this problem, we utilize existing coarse-grained datasets that offer a large number of annotations. A straightforward approach to address this problem is pre-finetuning, which employs coarse-grained data for representation learning. However, it cannot directly utilize the relationships between fine-grained and coarse-grained entities, although a fine-grained entity type is likely to be a subcategory of a coarse-grained entity type. We propose a fine-grained NER model with a Fine-to-Coarse(F2C) mapping matrix to leverage the hierarchical structure explicitly. In addition, we present an inconsistency filtering method to eliminate coarse-grained entities that are inconsistent with fine-grained entity types to avoid performance degradation. Our experimental results show that our method outperforms both $K$-shot learning and supervised learning methods when dealing with a small number of fine-grained annotations.

Recommender systems are important for providing personalized services to users, but the vast amount of collected user data has raised concerns about privacy (e.g., sensitive data), security (e.g., malicious data) and utility (e.g., toxic data). To address these challenges, recommendation unlearning has emerged as a promising approach, which allows specific data and models to be forgotten, mitigating the risks of sensitive/malicious/toxic user data. However, existing methods often struggle to balance completeness, utility, and efficiency, i.e., compromising one for the other, leading to suboptimal recommendation unlearning. In this paper, we propose an Interaction and Mapping Matrices Correction (IMCorrect) method for recommendation unlearning. Firstly, we reveal that many collaborative filtering (CF) algorithms can be formulated as mapping-based approach, in which the recommendation results can be obtained by multiplying the user-item interaction matrix with a mapping matrix. Then, IMCorrect can achieve efficient recommendation unlearning by correcting the interaction matrix and enhance the completeness and utility by correcting the mapping matrix, all without costly model retraining. Unlike existing methods, IMCorrect is a whitebox model that offers greater flexibility in handling various recommendation unlearning scenarios. Additionally, it has the unique capability of incrementally learning from new data, which further enhances its practicality. We conducted comprehensive experiments to validate the effectiveness of IMCorrect and the results demonstrate that IMCorrect is superior in completeness, utility, and efficiency, and is applicable in many recommendation unlearning scenarios.

Spatial signal processing algorithms often use pre-given coordinate systems to label pixel positions. These processing algorithms are thus burdened by an external reference grid, making the acquisition of relative, intrinsic features difficult. This is in contrast to animal vision and cognition: animals recognize features without an external coordinate system. We show that a coordinate system-independent algorithm for visual signal processing is not only important for animal vision, but also fundamental for concept formation. In this paper we start with a visual object deformation transfer experiment. We then formulate an algorithm that achieves deformation-invariance with relative coordinates. The paper concludes with implications for general concept formation.

Designing effective rehabilitation strategies for upper extremities, particularly hands and fingers, warrants the need for a computational model of human motor learning. The presence of large degrees of freedom (DoFs) available in these systems makes it difficult to balance the trade-off between learning the full dexterity and accomplishing manipulation goals. The motor learning literature argues that humans use motor synergies to reduce the dimension of control space. Using the low-dimensional space spanned by these synergies, we develop a computational model based on the internal model theory of motor control. We analyze the proposed model in terms of its convergence properties and fit it to the data collected from human experiments. We compare the performance of the fitted model to the experimental data and show that it captures human motor learning behavior well.

The phonological structure of human languages is intricate, yet highly constrained. Through a combination of connectionist modeling and linguistic analysis, we are attempting to develop a computational basis for the nature of phonology. We present a connectionist architecture that performs multiple simultaneous insertion, deletion, and mutation operations on sequences of phonemes, and introduce a novel additional primitive, clustering. Clustering provides an interesting alternative to both iterative and relaxation accounts of assimilation processes such as vowel harmony. Our resulting model is efficient because it processes utterances entirely in parallel using only feed-forward circuitry.

The phonological structure of human languages is intricate, yet highly constrained. Through a combination of connectionist modeling and linguistic analysis, we are attempting to develop a computational basis for the nature of phonology. We present a connectionist architecture that performs multiple simultaneous insertion, deletion, and mutation operations on sequences of phonemes, and introduce a novel additional primitive, clustering. Clustering provides an interesting alternative to both iterative and relaxation accounts of assimilation processes such as vowel harmony. Our resulting model is efficient because it processes utterances entirely in parallel using only feed-forward circuitry.

Through a combination linguistic analysis, we are attempting to develop a computational basis for the nature of phonology. We present a connectionist architecture that performs multiple simultaneous insertion, deletion, and mutation operations on sequences of phonemes, and introduce a novel additional primitive, clustering. Clustering provides an interesting alternative to both iterative and relaxation accounts of assimilation processes such as vowel harmony. Our resulting model is efficient because it processes utterances entirely in parallel using only feed-forward circuitry.