Prior research notes that BERT's computational cost grows quadratically with sequence length thus leading to longer training times, higher GPU memory constraints and carbon emissions. While recent work seeks to address these scalability issues at pre-training, these issues are also prominent in fine-tuning especially for long sequence tasks like document classification. Our work thus focuses on optimizing the computational cost of fine-tuning for document classification. We achieve this by complementary learning of both topic and language models in a unified framework, named TopicBERT. This significantly reduces the number of self-attention operations - a main performance bottleneck. Consequently, our model achieves a 1.4x ($\sim40\%$) speedup with $\sim40\%$ reduction in $CO_2$ emission while retaining $99.9\%$ performance over 5 datasets.
State-of-the-art reinforcement learning algorithms mostly rely on being allowed to directly interact with their environment to collect millions of observations. This makes it hard to transfer their success to industrial control problems, where simulations are often very costly or do not exist, and exploring in the real environment can potentially lead to catastrophic events. Recently developed, model-free, offline algorithms, can learn from a single dataset by mitigating extrapolation error in value functions. However, the robustness of the training process is still comparatively low, a problem known from methods using value functions. To improve robustness and stability of the learning process, we use dynamics models to assess policy performance instead of value functions, resulting in MOOSE (MOdel-based Offline policy Search with Ensembles), an algorithm which ensures low model bias by keeping the policy within the support of the data. We compare MOOSE with state-of-the-art model-free, offline RL algorithms BEAR and BCQ on the Industrial Benchmark and Mujoco continuous control tasks in terms of robust performance, and find that MOOSE outperforms its model-free counterparts in almost all considered cases, often even by far.
Lifelong learning has recently attracted attention in building machine learning systems that continually accumulate and transfer knowledge to help future learning. Unsupervised topic modeling has been popularly used to discover topics from document collections. However, the application of topic modeling is challenging due to data sparsity, e.g., in a small collection of (short) documents and thus, generate incoherent topics and sub-optimal document representations. To address the problem, we propose a lifelong learning framework for neural topic modeling that can continuously process streams of document collections, accumulate topics and guide future topic modeling tasks by knowledge transfer from several sources to better deal with the sparse data. In the lifelong process, we particularly investigate jointly: (1) sharing generative homologies (latent topics) over lifetime to transfer prior knowledge, and (2) minimizing catastrophic forgetting to retain the past learning via novel selective data augmentation, co-training and topic regularization approaches. Given a stream of document collections, we apply the proposed Lifelong Neural Topic Modeling (LNTM) framework in modeling three sparse document collections as future tasks and demonstrate improved performance quantified by perplexity, topic coherence and information retrieval task.
A crucial factor to trust Machine Learning (ML) algorithm decisions is a good representation of its application field by the training dataset. This is particularly true when parts of the training data have been artificially generated to overcome common training problems such as lack of data or imbalanced dataset. Over the last few years, Generative Adversarial Networks (GANs) have shown remarkable results in generating realistic data. However, this ML approach lacks an objective function to evaluate the quality of the generated data. Numerous GAN applications focus on generating image data mostly because they can be easily evaluated by a human eye. Less efforts have been made to generate time series data. Assessing their quality is more complicated, particularly for technical data. In this paper, we propose a human-centered approach supporting a ML or domain expert to accomplish this task using Visual Analytics (VA) techniques. The presented approach consists of two views, namely a GAN Iteration View showing similarity metrics between real and generated data over the iterations of the generation process and a Detailed Comparative View equipped with different time series visualizations such as TimeHistograms, to compare the generated data at different iteration steps. Starting from the GAN Iteration View, the user can choose suitable iteration steps for detailed inspection. We evaluate our approach with a usage scenario that enabled an efficient comparison of two different GAN models.
Topic models such as LDA, DocNADE, iDocNADEe have been popular in document analysis. However, the traditional topic models have several limitations including: (1) Bag-of-words (BoW) assumption, where they ignore word ordering, (2) Data sparsity, where the application of topic models is challenging due to limited word co-occurrences, leading to incoherent topics and (3) No Continuous Learning framework for topic learning in lifelong fashion, exploiting historical knowledge (or latent topics) and minimizing catastrophic forgetting. This thesis focuses on addressing the above challenges within neural topic modeling framework. We propose: (1) Contextualized topic model that combines a topic and a language model and introduces linguistic structures (such as word ordering, syntactic and semantic features, etc.) in topic modeling, (2) A novel lifelong learning mechanism into neural topic modeling framework to demonstrate continuous learning in sequential document collections and minimizing catastrophic forgetting. Additionally, we perform a selective data augmentation to alleviate the need for complete historical corpora during data hallucination or replay.
This paper describes our system (MIC-CIS) details and results of participation in the fine-grained propaganda detection shared task 2019. To address the tasks of sentence (SLC) and fragment level (FLC) propaganda detection, we explore different neural architectures (e.g., CNN, LSTM-CRF and BERT) and extract linguistic (e.g., part-of-speech, named entity, readability, sentiment, emotion, etc.), layout and topical features. Specifically, we have designed multi-granularity and multi-tasking neural architectures to jointly perform both the sentence and fragment level propaganda detection. Additionally, we investigate different ensemble schemes such as majority-voting, relax-voting, etc. to boost overall system performance. Compared to the other participating systems, our submissions are ranked 3rd and 4th in FLC and SLC tasks, respectively.
In this paper, we present a Bayesian view on model-based reinforcement learning. We use expert knowledge to impose structure on the transition model and present an efficient learning scheme based on variational inference. This scheme is applied to a heteroskedastic and bimodal benchmark problem on which we compare our results to NFQ and show how our approach yields human-interpretable insight about the underlying dynamics while also increasing data-efficiency.
We propose a novel Bayesian approach to modelling multimodal data generated by multiple independent processes, simultaneously solving the data association and induced supervised learning problems. Underpinning our approach is the use of Gaussian process priors which encode structure both on the functions and the associations themselves. The association of samples and functions are determined by taking both inputs and outputs into account while also obtaining a posterior belief about the relevance of the global components throughout the input space. We present an efficient learning scheme based on doubly stochastic variational inference and discuss how it can be applied to deep Gaussian process priors. We show results for an artificial data set, a noise separation problem, and a multimodal regression problem based on the cart-pole benchmark.
Past work in relation extraction mostly focuses on binary relation between entity pairs within single sentence. Recently, the NLP community has gained interest in relation extraction in entity pairs spanning multiple sentences. In this paper, we propose a novel architecture for this task: inter-sentential dependency-based neural networks (iDepNN). iDepNN models the shortest and augmented dependency paths via recurrent and recursive neural networks to extract relationships within (intra-) and across (inter-) sentence boundaries. Compared to SVM and neural network baselines, iDepNN is more robust to false positives in relationships spanning sentences. We evaluate our models on four datasets from newswire (MUC6) and medical (BioNLP shared task) domains that achieve state-of-the-art performance and show a better balance in precision and recall for inter-sentential relationships. We perform better than 11 teams participating in the BioNLP shared task 2016 and achieve a gain of 5.2% (0.587 vs 0.558) in F1 over the winning team. We also release the crosssentence annotations for MUC6.