Maximum likelihood estimation (MLE) is the predominant algorithm for training text generation models. This paradigm relies on direct supervision examples, which is not applicable to many emerging applications, such as generating adversarial attacks or generating prompts to control language models. Reinforcement learning (RL) on the other hand offers a more flexible solution by allowing users to plug in arbitrary task metrics as reward. Yet previous RL algorithms for text generation, such as policy gradient (on-policy RL) and Q-learning (off-policy RL), are often notoriously inefficient or unstable to train due to the large sequence space and the sparse reward received only at the end of sequences. In this paper, we introduce a new RL formulation for text generation from the soft Q-learning (SQL) perspective. It enables us to draw from the latest RL advances, such as path consistency learning, to combine the best of on-/off-policy updates, and learn effectively from sparse reward. We apply the approach to a wide range of novel text generation tasks, including learning from noisy/negative examples, adversarial attacks, and prompt generation. Experiments show our approach consistently outperforms both task-specialized algorithms and the previous RL methods.

Knowledge underpins reasoning. Recent research demonstrates that when relevant knowledge is provided as additional context to commonsense question answering (QA), it can substantially enhance the performance even on top of state-of-the-art. The fundamental challenge is where and how to find such knowledge that is high quality and on point with respect to the question; knowledge retrieved from knowledge bases are incomplete and knowledge generated from language models are inconsistent. We present Rainier, or Reinforced Knowledge Introspector, that learns to generate contextually relevant knowledge in response to given questions. Our approach starts by imitating knowledge generated by GPT-3, then learns to generate its own knowledge via reinforcement learning where rewards are shaped based on the increased performance on the resulting question answering. Rainier demonstrates substantial and consistent performance gains when tested over 9 different commonsense benchmarks: including 5 datasets that are seen during model training, as well as 4 datasets that are kept unseen. Our work is the first to report that knowledge generated by models that are orders of magnitude smaller than GPT-3, even without direct supervision on the knowledge itself, can exceed the quality of commonsense knowledge elicited from GPT-3.

We make an attempt to understanding convolutional neural network by exploring the relationship between (deep) convolutional neural networks and Volterra convolutions. We propose a novel approach to explain and study the overall characteristics of neural networks without being disturbed by the horribly complex architectures. Specifically, we attempt to convert the basic structures of a convolutional neural network (CNN) and their combinations to the form of Volterra convolutions. The results show that most of convolutional neural networks can be approximated in the form of Volterra convolution, where the approximated proxy kernels preserve the characteristics of the original network. Analyzing these proxy kernels may give valuable insight about the original network. Base on this setup, we presented methods to approximating the order-zero and order-one proxy kernels, and verified the correctness and effectiveness of our results.

A number of models have been developed in India to forecast the spread of the coronavirus disease or COVID-19 in the country. Model building has had to incorporate our evolving knowledge of the disease, including the appearance of new variants, immune escape leading to reinfections, time-varying non-pharmaceutical interventions, the pace of the vaccination program, and breakthrough infections. The predictive power of these models has been hampered by the lack of availability of quality data on infection and deaths as a function of age, the nature of social contacts, demography, and the clinical consequence of infection. An early emphasis on "ensemble models," a thrust toward increased data availability, a greater engagement of modelers with the epidemiological and public health communities, and a more nuanced approach to communicating the limitations of modeling could have substantially increased the usefulness of models during the COVID-19 pandemic in India. Most models were variants of the SEIR model where the individuals in the population move from S susceptible to E exposed to I infectious to R removed compartments.

We use deep neural networks to machine learn correlations between knot invariants in various dimensions. The three-dimensional invariant of interest is the Jones polynomial $J(q)$, and the four-dimensional invariants are the Khovanov polynomial $\text{Kh}(q,t)$, smooth slice genus $g$, and Rasmussen's $s$-invariant. We find that a two-layer feed-forward neural network can predict $s$ from $\text{Kh}(q,-q^{-4})$ with greater than $99\%$ accuracy. A theoretical explanation for this performance exists in knot theory via the now disproven knight move conjecture, which is obeyed by all knots in our dataset. More surprisingly, we find similar performance for the prediction of $s$ from $\text{Kh}(q,-q^{-2})$, which suggests a novel relationship between the Khovanov and Lee homology theories of a knot. The network predicts $g$ from $\text{Kh}(q,t)$ with similarly high accuracy, and we discuss the extent to which the machine is learning $s$ as opposed to $g$, since there is a general inequality $|s| \leq 2g$. The Jones polynomial, as a three-dimensional invariant, is not obviously related to $s$ or $g$, but the network achieves greater than $95\%$ accuracy in predicting either from $J(q)$. Moreover, similar accuracy can be achieved by evaluating $J(q)$ at roots of unity. This suggests a relationship with $SU(2)$ Chern--Simons theory, and we review the gauge theory construction of Khovanov homology which may be relevant for explaining the network's performance.

Named geographic entities (geo-entities for short) are the building blocks of many geographic datasets. Characterizing geo-entities is integral to various application domains, such as geo-intelligence and map comprehension, while a key challenge is to capture the spatial-varying context of an entity. We hypothesize that we shall know the characteristics of a geo-entity by its surrounding entities, similar to knowing word meanings by their linguistic context. Accordingly, we propose a novel spatial language model, SpaBERT, which provides a general-purpose geo-entity representation based on neighboring entities in geospatial data. SpaBERT extends BERT to capture linearized spatial context, while incorporating a spatial coordinate embedding mechanism to preserve spatial relations of entities in the 2-dimensional space. SpaBERT is pretrained with masked language modeling and masked entity prediction tasks to learn spatial dependencies. We apply SpaBERT to two downstream tasks: geo-entity typing and geo-entity linking. Compared with the existing language models that do not use spatial context, SpaBERT shows significant performance improvement on both tasks. We also analyze the entity representation from SpaBERT in various settings and the effect of spatial coordinate embedding.

However, the previous methods, including fixed Simultaneous machine translation (SiMT) (Gu and adaptive policies, lack evaluation before taking et al., 2017; Ma et al., 2019; Arivazhagan et al., the next action. For fixed policy (Ma et al., 2019; 2019; Ma et al., 2020; Zhang and Feng, 2021b, Elbayad et al., 2020; Zhang et al., 2021; Zhang 2022d) starts translation before reading the whole and Feng, 2021c), the model generates translation source sentence. It seeks to achieve good latencyquality according to the predefined translation rules. Although tradeoffs and is suitable for various scenarios it only relies on simple training methods, with different latency tolerances. Compared to it cannot make full use of the context to decide an full-sentence machine translation, SiMT is more appropriate translation policy. For adaptive policy challenging because it lacks partial source content (Gu et al., 2017; Arivazhagan et al., 2019; Ma in translation and needs to decide on translation et al., 2020; Zhang et al., 2022), the model can policy additionally.

Remote sensing is a higher technology to produce knowledge for data mining applications. In principle hyperspectral images (HSIs) is a remote sensing tool that provides precise classification of regions. The HSI contains more than a hundred of images of the ground truth (GT) map. Some images are carrying relevant information, but others describe redundant information, or they are affected by atmospheric noise. The aim is to reduce dimensionality of HSI. Many studies use mutual information (MI) or normalised forms of MI to select appropriate bands. In this paper we design an algorithm based also on MI, and we combine MI with steepest ascent algorithm, to improve a symmetric uncertainty coefficient-based strategy to select relevant bands for classification of HSI. This algorithm is a feature selection tool and a wrapper strategy. We perform our study on HSI AVIRIS 92AV3C. This is an artificial intelligent system to control redundancy; we had to clear the difference of the result's algorithm and the human decision, and this can be viewed as case study which human decision is perhaps different to an intelligent algorithm. Index Terms - Hyperspectral images, Classification, Fea-ture selection, Mutual Information, Redundancy, Steepest Ascent. Artificial Intelligence

Conventional autoregressive left-to-right (L2R) sequence generation faces two issues during decoding: limited to unidirectional target sequence modeling, and constrained on strong local dependencies. To address the aforementioned problem, we propose P$^3$LM, a probabilistically permuted prophet language model, which strengthens the modeling of bidirectional information and long token dependencies for sequence generation. Specifically, P$^3$LM learns to generate tokens in permuted order upon an order-aware transformer decoder, as well as to generate the corresponding future $N$ tokens with a multi-stream attention mechanism. Extensive experiments are conducted on the GLGE benchmark, which includes four datasets for summarization, two for question generation, one for conversational question answering, and one for dialog response generation, where P$^3$LM achieves state-of-the-art results compared with strong publicly available generative pre-training methods.

Contrastive learning (CL) brought significant progress to various NLP tasks. Despite this progress, CL has not been applied to Arabic NLP to date. Nor is it clear how much benefits it could bring to particular classes of tasks such as those involved in Arabic social meaning (e.g., sentiment analysis, dialect identification, hate speech detection). In this work, we present a comprehensive benchmark study of state-of-the-art supervised CL methods on a wide array of Arabic social meaning tasks. Through extensive empirical analyses, we show that CL methods outperform vanilla finetuning on most tasks we consider. We also show that CL can be data efficient and quantify this efficiency. Overall, our work allows us to demonstrate the promise of CL methods, including in low-resource settings.