Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userTruth-Conditional Captions for Time Series Data
التعليق الشرطي للحقيقة لبيانات سلسلة الوقت
118
0 0
0.0
(
0
)
Created by
Shamra Editor
Ask ChatGPT about the research
في هذه الورقة، نستكشف مهمة توليد أوصاف اللغة الطبيعية تلقائيا لأنماط بارزة في سلسلة زمنية، مثل أسعار الأسهم لشركة أكثر من أسبوع. يجب أن يكون نموذج لهذه المهمة قادرا على استخراج أنماط رفيعة المستوى مثل وجود ذروة أو تراجع. في حين أن النماذج العصبية المعاصرة النموذجية مع آليات الاهتمام يمكن أن تولد أوصاف إخراج بطلاقة لهذه المهمة، فإنها غالبا ما تولد أوصاف غير صحيحة في الواقع. نقترح نموذجا حسابيا مع بنية شرائط للحقيقة تعمل أولا البرامج المستفادة الصغيرة على سلسلة وقت الإدخال، ثم يحدد البرامج / الأنماط التي تمسك بالإدخال المحدد، وأخيرا ظروف * فقط * البرنامج الصحيح الذي تم اختياره (بدلا من ذلك من سلسلة وقت الإدخال) لتوليد وصف نص الإخراج. يتم إنشاء برنامج في طرازنا من الوحدات النمطية، وهي شبكات عصبية صغيرة مصممة لالتقاط الأنماط العددية والمعلومات الزمنية. يتم تقاسم الوحدات النمطية عبر برامج متعددة، مما يتيح التركيبية وكذلك التعلم الفعال لمعلمات الوحدة النمطية. إن الوحدات النمطية، وكذلك تكوين الوحدات النمطية، غير مقصودة في البيانات، ونحن نتعلمهم في أزياء نهاية إلى نهاية مع إشارة التدريب الوحيدة القادمة من أوصاف نص اللغة الطبيعية المصاحبة. نجد أن النموذج المقترح قادر على توليد التسميات التوضيحية عالية الدقة على الرغم من أننا نعتبر مساحة صغيرة وبسيطة من أنواع الوحدات النمطية.
In this paper, we explore the task of automatically generating natural language descriptions of salient patterns in a time series, such as stock prices of a company over a week. A model for this task should be able to extract high-level patterns such as presence of a peak or a dip. While typical contemporary neural models with attention mechanisms can generate fluent output descriptions for this task, they often generate factually incorrect descriptions. We propose a computational model with a truth-conditional architecture which first runs small learned programs on the input time series, then identifies the programs/patterns which hold true for the given input, and finally conditions on *only* the chosen valid program (rather than the input time series) to generate the output text description. A program in our model is constructed from modules, which are small neural networks that are designed to capture numerical patterns and temporal information. The modules are shared across multiple programs, enabling compositionality as well as efficient learning of module parameters. The modules, as well as the composition of the modules, are unobserved in data, and we learn them in an end-to-end fashion with the only training signal coming from the accompanying natural language text descriptions. We find that the proposed model is able to generate high-precision captions even though we consider a small and simple space of module types.
References used
https://aclanthology.org/
rate research
Read More
We propose a rolling version of the Latent Dirichlet Allocation, called RollingLDA. By a sequential approach, it enables the construction of LDA-based time series of topics that are consistent with previous states of LDA models. After an initial mode
ling, updates can be computed efficiently, allowing for real-time monitoring and detection of events or structural breaks. For this purpose, we propose suitable similarity measures for topics and provide simulation evidence of superiority over other commonly used approaches. The adequacy of the resulting method is illustrated by an application to an example corpus. In particular, we compute the similarity of sequentially obtained topic and word distributions over consecutive time periods. For a representative example corpus consisting of The New York Times articles from 1980 to 2020, we analyze the effect of several tuning parameter choices and we run the RollingLDA method on the full dataset of approximately 4 million articles to demonstrate its feasibility.
Probing experiments investigate the extent to which neural representations make properties---like part-of-speech---predictable. One suggests that a representation encodes a property if probing that representation produces higher accuracy than probing
a baseline representation like non-contextual word embeddings. Instead of using baselines as a point of comparison, we're interested in measuring information that is contained in the representation but not in the baseline. For example, current methods can detect when a representation is more useful than the word identity (a baseline) for predicting part-of-speech; however, they cannot detect when the representation is predictive of just the aspects of part-of-speech not explainable by the word identity. In this work, we extend a theory of usable information called V-information and propose conditional probing, which explicitly conditions on the information in the baseline. In a case study, we find that after conditioning on non-contextual word embeddings, properties like part-of-speech are accessible at deeper layers of a network than previously thought.
How to generate summaries of different styles without requiring corpora in the target styles, or training separate models? We present two novel methods that can be deployed during summary decoding on any pre-trained Transformer-based summarization mo
del. (1) Decoder state adjustment instantly modifies decoder final states with externally trained style scorers, to iteratively refine the output against a target style. (2) Word unit prediction constrains the word usage to impose strong lexical control during generation. In experiments of summarizing with simplicity control, automatic evaluation and human judges both find our models producing outputs in simpler languages while still informative. We also generate news headlines with various ideological leanings, which can be distinguished by humans with a reasonable probability.
حظيت نمذجة وتوقع السلاسل الزمنية بأهمية كبيرة في العديد من المجالات التطبيقية كالتنبؤ بالطقس وأسعار العملات ومعدلات استهلاك الوقود والكهرباء، إن توقع السلاسل الزمنية من شأنه أن يزود المنظمات والشركات بالمعلومات الضرورية لاتخاذ القرارات الهامة، وبسبب
أهمية هذا المجال من الناحية التطبيقية فإن الكثير من الأعمال البحثية التي جرت ضمنه خلال السنوات الماضية، إضافةً إلى العدد الكبير من النماذج والخوارزميات التي تم اقتراحها في أدب البحث العلمي والتي كان هدفها تحسين كل من الدقة والكفاءة في نمذجة وتوقع السلاسل الزمنية.
Deriving and modifying graphs from natural language text has become a versatile basis technology for information extraction with applications in many subfields, such as semantic parsing or knowledge graph construction. A recent work used this techniq
ue for modifying scene graphs (He et al. 2020), by first encoding the original graph and then generating the modified one based on this encoding. In this work, we show that we can considerably increase performance on this problem by phrasing it as graph extension instead of graph generation. We propose the first model for the resulting graph extension problem based on autoregressive sequence labelling. On three scene graph modification data sets, this formulation leads to improvements in accuracy over the state-of-the-art between 13 and 24 percentage points. Furthermore, we introduce a novel data set from the biomedical domain which has much larger linguistic variability and more complex graphs than the scene graph modification data sets. For this data set, the state-of-the art fails to generalize, while our model can produce meaningful predictions.
suggested questions
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
