Dialog response generation in open domain is an important research topic where the main challenge is to generate relevant and diverse responses. In this paper, we propose a new dialog pre-training framework called DialogVED, which introduces continuous latent variables into the enhanced encoder-decoder pre-training framework to increase the relevance and diversity of responses. With the help of a large dialog corpus (Reddit), we pre-train the model using the following 4 tasks adopted in language models (LMs) and variational autoencoders (VAEs): 1) masked language model; 2) response generation; 3) bag-of-words prediction; and 4) KL divergence reduction. We also add additional parameters to model the turn structure in dialogs to improve the performance of the pre-trained model. We conduct experiments on PersonaChat, DailyDialog, and DSTC7-AVSD benchmarks for response generation. Experimental results show that our model achieves the new state-of-the-art results on all these datasets.

Picture this: A newborn baby gets thrown into a deep-end pool and needs to learn how to swim for survival. This newborn baby will go through the method of trial and error to improve itself. It will learn from its mistakes and improve its accuracy over time. An artificial neural network (ANN) is a computing system that can learn on its own. It creates an adaptive system that computers use to learn from their mistakes and improve continuously.

Research in the field of machine learning and AI, now a key technology in practically every industry and company, is far too voluminous for anyone to read it all. This column, Perceptron, aims to collect some of the most relevant recent discoveries and papers -- particularly in, but not limited to, artificial intelligence -- and explain why they matter. Over the past few weeks, researchers at MIT have detailed their work on a system to track the progression of Parkinson's patients by continuously monitoring their gait speed. Elsewhere, Whale Safe, a project spearheaded by the Benioff Ocean Science Laboratory and partners, launched buoys equipped with AI-powered sensors in an experiment to prevent ships from striking whales. Other aspects of ecology and academics also saw advances powered by machine learning.

In this paper, we examine the use of data from multiple sensing modes, i.e., accelerometry and global navigation satellite system (GNSS), for classifying animal behavior. We extract three new features from the GNSS data, namely, distance from water point, median speed, and median estimated horizontal position error. We combine the information available from the accelerometry and GNSS data via two approaches. The first approach is based on concatenating the features extracted from both sensor data and feeding the concatenated feature vector into a multi-layer perceptron (MLP) classifier. The second approach is based on fusing the posterior probabilities predicted by two MLP classifiers. The input to each classifier is the features extracted from the data of one sensing mode. We evaluate the performance of the developed multimodal animal behavior classification algorithms using two real-world datasets collected via smart cattle collar tags and ear tags. The leave-one-animal-out cross-validation results show that both approaches improve the classification performance appreciably compared with using data of only one sensing mode. This is more notable for the infrequent but important behaviors of walking and drinking. The algorithms developed based on both approaches require little computational and memory resources hence are suitable for implementation on embedded systems of our collar tags and ear tags. However, the multimodal animal behavior classification algorithm based on posterior probability fusion is preferable to the one based on feature concatenation as it delivers better classification accuracy, has less computational and memory complexity, is more robust to sensor data failure, and enjoys better modularity.

Prediction of the future performance of academic journals is a task that can benefit a variety of stakeholders including editorial staff, publishers, indexing services, researchers, university administrators and granting agencies. Using historical data on journal performance, this can be framed as a machine learning regression problem. In this work, we study two such regression tasks: 1) prediction of the number of citations a journal will receive during the next calendar year, and 2) prediction of the Elsevier CiteScore a journal will be assigned for the next calendar year. To address these tasks, we first create a dataset of historical bibliometric data for journals indexed in Scopus. We propose the use of neural network models trained on our dataset to predict the future performance of journals. To this end, we perform feature selection and model configuration for a Multi-Layer Perceptron and a Long Short-Term Memory. Through experimental comparisons to heuristic prediction baselines and classical machine learning models, we demonstrate superior performance in our proposed models for the prediction of future citation and CiteScore values.

In the previous post about TabTransformer I've described how the model works and how it can be applied to your data. This post will build on it, so if you haven't read it yet, I highly recommend starting there and returning to this post afterwards. TabTransformer was shown to outperform traditional multi-layer perceptrons (MLPs) and came close to the performance of Gradient Boosted Trees (GBTs) on some datasets. However, there is one noticeable drawback with the architecture -- it doesn't take numerical features into account when constructing contextual embeddings. This post deep dives into the paper by Gorishniy et al. (2021) which has addressed this issue by introducing FT-Transformer (Feature Tokenizer Transformer).

The number of clinical citations received from clinical guidelines or clinical trials has been considered as one of the most appropriate indicators for quantifying the clinical impact of biomedical papers. Therefore, the early prediction of the clinical citation count of biomedical papers is critical to scientific activities in biomedicine, such as research evaluation, resource allocation, and clinical translation. In this study, we designed a four-layer multilayer perceptron neural network (MPNN) model to predict the clinical citation count of biomedical papers in the future by using 9,822,620 biomedical papers published from 1985 to 2005. We extracted ninety-one paper features from three dimensions as the input of the model, including twenty-one features in the paper dimension, thirty-five in the reference dimension, and thirty-five in the citing paper dimension. In each dimension, the features can be classified into three categories, i.e., the citation-related features, the clinical translation-related features, and the topic-related features. Besides, in the paper dimension, we also considered the features that have previously been demonstrated to be related to the citation counts of research papers. The results showed that the proposed MPNN model outperformed the other five baseline models, and the features in the reference dimension were the most important.

The message-passing mechanism helps Graph Neural Networks (GNNs) achieve remarkable results on various node classification tasks. Nevertheless, the recursive nodes fetching and aggregation in message-passing cause inference latency when deploying GNNs to large-scale graphs. One promising inference acceleration direction is to distill the GNNs into message-passing-free student multi-layer perceptrons (MLPs). However, the MLP student cannot fully learn the structure knowledge due to the lack of structure inputs, which causes inferior performance in the heterophily and inductive scenarios. To address this, we intend to inject structure information into MLP-like students in low-latency and interpretable ways. Specifically, we first design a Structure-Aware MLP (SA-MLP) student that encodes both features and structures without message-passing. Then, we introduce a novel structure-mixing knowledge distillation strategy to enhance the learning ability of MLPs for structure information. Furthermore, we design a latent structure embedding approximation technique with two-stage distillation for inductive scenarios. Extensive experiments on eight benchmark datasets under both transductive and inductive settings show that our SA-MLP can consistently outperform the teacher GNNs, while maintaining faster inference as MLPs. The source code of our work can be found in https://github.com/JC-202/SA-MLP.

The classical Perceptron algorithm of Rosenblatt can be used to find a linear threshold function to correctly classify $n$ linearly separable data points, assuming the classes are separated by some margin $\gamma > 0$. A foundational result is that Perceptron converges after $\Omega(1/\gamma^{2})$ iterations. There have been several recent works that managed to improve this rate by a quadratic factor, to $\Omega(\sqrt{\log n}/\gamma)$, with more sophisticated algorithms. In this paper, we unify these existing results under one framework by showing that they can all be described through the lens of solving min-max problems using modern acceleration techniques, mainly through optimistic online learning. We then show that the proposed framework also lead to improved results for a series of problems beyond the standard Perceptron setting. Specifically, a) For the margin maximization problem, we improve the state-of-the-art result from $O(\log t/t^2)$ to $O(1/t^2)$, where $t$ is the number of iterations; b) We provide the first result on identifying the implicit bias property of the classical Nesterov's accelerated gradient descent (NAG) algorithm, and show NAG can maximize the margin with an $O(1/t^2)$ rate; c) For the classical $p$-norm Perceptron problem, we provide an algorithm with $\Omega(\sqrt{(p-1)\log n}/\gamma)$ convergence rate, while existing algorithms suffer the $\Omega({(p-1)}/\gamma^2)$ convergence rate.

Previous works on Treatment Effect Estimation (TEE) are not in widespread use because they are predominantly theoretical, where strong parametric assumptions are made but untractable for practical application. Recent work uses multilayer perceptron (MLP) for modeling casual relationships, however, MLPs lag far behind recent advances in ML methodology, which limits their applicability and generalizability. To extend beyond the single domain formulation and towards more realistic learning scenarios, we explore model design spaces beyond MLPs, i.e., transformer backbones, which provide flexibility where attention layers govern interactions among treatments and covariates to exploit structural similarities of potential outcomes for confounding control. Through careful model design, Transformers as Treatment Effect Estimators (TransTEE) is proposed. We show empirically that TransTEE can: (1) serve as a general purpose treatment effect estimator that significantly outperforms competitive baselines in a variety of challenging TEE problems (e.g., discrete, continuous, structured, or dosage-associated treatments) and is applicable to both when covariates are tabular and when they consist of structural data (e.g., texts, graphs); (2) yield multiple advantages: compatibility with propensity score modeling, parameter efficiency, robustness to continuous treatment value distribution shifts, explainable in covariate adjustment, and real-world utility in auditing pre-trained language models