Paper Review

please modify the following [TOC] Title: LTL2Action: Generalizing LTL Instructions for Multi-Task RL Author: Pashootan Vaezipoor et. al. Publish Year: 2021 Review Date: March 2022 Summary of paper Motivation they addressed the problem of teaching a deep reinforcement learning agent to follow instructions in multi-task environments. Instructions are expressed in a well-known formal language – linear temporal logic (LTL) Limitation of the vanilla MDP temporal constraints cannot be expressed as rewards in MDP setting and thus modular policy and other stuffs are not able to obtain maximum rewards. ...

[TOC] Title: Learning to Ground Language to Temporal Logical Form Author: Roma Patel et. al. Publish Year: 2019 Review Date: Feb 2022 Summary of paper Motivation natural language commands often exhibits sequential (temporal) constraints e.g., “go through the kitchen and then into the living room”. But this constraints cannot be expressed in the reward of Markov Decision Process setting. (see this paper) Therefore, they proposed to ground language to Linear Temporal logic (LTL) and after that continue to map from LTL expressions to action sequences. ...

[TOC] Title: Out of Order: How Important Is The Sequential Order of Words in a Sentence in Natural Language Understanding Tasks? Author: Thang M. Pham Publish Year: Jul 2021 Review Date: Feb 2022 Summary of paper The author found out that BERT-based models trained on GLUE have low sensitivity to word orders. The research questions are the following Do BERT-based models trained on GLUE care about the order of words in a sentence? ANS: NO, except one task named CoLA, which is to detecting grammatically incorrect sentences. Surprisingly, for the rest of the 5 out of 6 binary-classification tasks (i.e. except CoLA), between75% and 90% of the originally correct predictions remain constant after 1-grams are randomly re-ordered Are SOTA BERT-based models using word order information when solving NLU tasks? If not, what cues do they rely on? ANS: they heavily rely on the word itself rather than the ordering. The results showed that if the top - 1 most important word measured by LIME has a positive meaning, then there is 100% probability that the sentence’s label is “positive” Results ...

[TOC] Title: Guided K-best Selection for Semantic Parsing Annotation Author: Anton Belyy et. al. Publish Year: 2021 Review Date: Feb 2022 Summary of paper Motivation They wanted to tackle the challenge of efficient data collection (data annotation) for the conversational semantic parsing task. In the presence of little available training data, they proposed human-in-the-loop interfaces for guided K-best selection, using a prototype model trained on limited data. Result Their user studies showed that the keyword searching function combined with a keyword suggestion method strikes the balance between annotation accuracy and speed ...

[TOC] Title: Compositionality as Lexical Symmetry Author: Ekin Akyurek; Jacob Andreas Publish Year: Jan 2022 Review Date: Feb 2022 Summary of paper Motivation Standard deep network models lack the inductive bias needed to generalize compositionally in tasks like semantic parsing, translation, and question answering. So, a large body of work in NLP seeks to overcome this limitation with new model architectures that enforce a compositional process of sentence interpretation. Goal ...

[TOC] Title: When Attention Meets Fast Recurrence: Training Language Models with Reduce Compute Author: Tao Lei Publish Year: Sep 2021 Review Date: Jan 2022 Summary of paper As the author mentioned, the inspiration of SRU++ comes from two lines of research: paralleization / speed problem of Original RNN leveraging recurrence in conjunction with self-attention Structure of SRU++ New discovery :little attention is needed given recurrence. Similar to the observation of Merity (2019), they found using a couple of attention layers sufficient to obtain SOTA results. ...

[TOC] Title: GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models Author: Alex Nichol et. al. Publish Year: Dec 2021 Review Date: Jan 2022 Summary of paper In author’s previous work, the diffusion model can achieve photorealism in the class-conditional setting by augmenting with classifier guidance, a technique which allows diffusion models to condition on a classifier’s labels. The classifier is first trained on noised images, and during the diffusion sampling process, gradients from the classifier are used to guide the output sample towards the label. classifier details ...

[TOC] Title: M6: A Chinese Multimodal Pretrainer Author: Junyang Lin et. al. Publish Year: May 2021 Review Date: Jan 2022 Summary of paper This paper re-emphasises that large model trained on big data have extremely large capacity and it can outperform the SOTA in downstream tasks especially in the zero-shot setting. So, the author trained a big multi-modal model Also, they proposed a innovative way to tackle downstream tasks. they use masks to block cross attention between tokens so as to fit different types of downstream task Key idea: mask tokens during cross attention so as to solve certain tasks Overview ...

[TOC] Title: BABYAI++: Towards Grounded-Language Learning Beyond Memorization Author: Tianshi Cao et. al. Publish Year: 2020 ICLR Review Date: Jan 2022 Summary of paper The paper introduced a new RL environment BabyAI++ that can investigate whether RL agents can extract knowledge from descriptive text and eventually increase generalisation performance. BabyAI++ environment example the descriptive text describe the feature of the object. notice that the feature of object can easily change as we change the descriptive text. Model ...

[TOC] Title: Language in a (Search) Box: Grounding Language Learning in Real-World Human-Machine Interaction Author: Federico Bianchi Publish Year: 2021 Review Date: Jan 2022 Summary of paper the author investigated grounded language learning through the natural interaction between users and the shopping website search engine. How they do it convert the shopping object dataset into a Latent Grounded Domain related products end up closer in the embedding space train the mapping model (mapping from text query to a portion of product space) based on the user click behaviour (In the training dataset, the users queries about “Nike” and the they would click relevant Nike Product) ...

[TOC] Title: Decision Transformer: Reinforcement Learning via Sequence Modeling Author: Lili Chen et. al. Publish Year: Jun 2021 Review Date: Dec 2021 Summary of paper The Architecture of Decision Transformer Inputs are reward, observation and action Outputs are action, in training time, the future action will be masked out. I believe this model is able to generate a very good long sequence of actions due to transformer architecture. But somehow this is not RL anymore because the transformer is not trained by reward signal … ...

[TOC] Title: Grammar-Based Grounded Lexicon Learning Author: Jiayuan Mao Publish Year: 2021 NeurIPS Review Date: Dec 2021 Summary of paper The paper extend the previous work “Neuro-Symbolic Concept Learner” by parsing the natural language questions using symbolic manner. The core semantic parsing technique is Combinatory Categorical Grammar with CKY algorithm to prune unlikely expressions. The full picture looks like this The detailed algorithm process looks like this How to derive concept embedding ...

[TOC] Title: Interactive Grounded Language Understanding in a Collaborative Environment Author: Julia Kiseleva et. al. Publish Year: 2021 Review Date: Dec 2021 Summary of paper The primary goal of the competition is to approach the problem of how to build interactive agents that learn to solve a task while provided with grounded natural language instructions in a collaborative environment. The split the problem into following concrete research questions, which correspond to separate tasks that can be used to study each component individually before joining all of them into one system ...

[TOC] Title: Expressing and Exploiting the Common Subgoal Structure of Classical Planning Domains Using Sketches Author: Dominik Drexler et. al. Publish Year: 2021 Review Date: Dec 2021 Summary of paper Algorithms like SIW often fail when the goal is not easily serialisable or when some of the subproblems have a high width. In this work, the author address these limitations by using a simple but powerful language for expressing finer problem decompositions called policy sketches. ...

[TOC] Title: Language as an Abstraction for Hierarchical Deep Reinforcement Learning Author: Yiding Jiang et. al. Publish Year: 2019 NeurIPS Review Date: Dec 2021 Summary of paper Solving complex, temporally-extended tasks is a long-standing problem in RL. Acquiring effective yet general abstractions for hierarchical RL is remarkably challenging. Therefore, they propose to use language as the abstraction, as it provides unique compositional structure, enabling fast learning and combinatorial generalisation ...

[TOC] Title: Hierarchical Decision Making by Generating and Following Natural Language Instructions Author: Hengyuan Hu et. al. FAIR Publish Year: 2019 Review Date: Dec 2021 Summary of paper One line summary: they build a Architect Builder model to clone human behaviour for playing RTS game Their task environment is very similar to IGLU competition setting, but their model is too task-specific The author mentioned some properties about natural language instructions ...

Title: Attention Over Learned Object Embeddings Enables Complex Visual Reasoning Author: David Ding et. al. Publish Year: 2021 NeurIPS Review Date: Dec 2021 Background info for this paper: Their paper propose a all-in-one transformer model that is able to answer CLEVRER counterfactual questions with higher accuracy (75.6% vs 46.5%) and less training data (- 40%) They believe that their model relies on three key aspects: self-attention soft-discretization self-supervised learning ...

Title: Modular Multitask Reinforcement Learning with Policy Sketches Author: Jacob Andreas et. al. Publish Year: 2017 Review Date: Dec 2021 Background info for this paper: Their paper describe a framework that is inspired by on options MDP, for which a reinforcement learning task is handled by several sub-MDP modules. (that is why they call it Modular RL) They consider a multitask RL problem in a shared environment. (See the figure below). The IGLU Minecraft challenge as well as Angry Birds also belongs to this category. ...

[TOC] Title: On the Expressivity of Markov Reward Author: David Abel et. al. Publish Year: NuerIPS 2021 Review Date: 6 Dec 2021 Summary of paper This needs to be only 1-3 sentences, but it demonstrates that you understand the paper and, moreover, can summarize it more concisely than the author in his abstract. The author found out that in the Markov Decision Process scenario, (i.e., we do not look at the history of the trajectory to provide rewards), some tasks cannot be realised perfectly by reward functions. i.e., ...

[TOC] Title: Deep Reinforcement Learning at the Edge of the Statistical Precipice Author: Rishabh Agarwal et. al. Publish Year: NeurIPS 2021 Review Date: 3 Dec 2021 Summary of paper This needs to be only 1-3 sentences, but it demonstrates that you understand the paper and, moreover, can summarize it more concisely than the author in his abstract. Most current published results on deep RL benchmarks uses point estimate of aggregate performance such as mean and median score across the task. ...

[TOC] Title: Reward learning from human preferences and demonstractions in Atari Author: Borja Ibarz et. al. Publish Year: 2018 Review Date: Nov 2021 Summary of paper This needs to be only 1-3 sentences, but it demonstrates that you understand the paper and, moreover, can summarize it more concisely than the author in his abstract. The author proposed a method that uses human expert’s annotation rather than extrinsic reward from the environment to guide the reinforcement learning. ...

[TOC] Title: Go-Explore: a New Approach for Hard-Exploration Problems Author: Adrien Ecoffet et. al. Publish Year: 2021 Review Date: Nov 2021 Summary of paper This needs to be only 1-3 sentences, but it demonstrates that you understand the paper and, moreover, can summarize it more concisely than the author in his abstract. The author hypothesised that there are two main issues that prevent DRL agents from achieving high score in exploration-hard game (e.g., Montezuma’s Revenge) ...

[TOC] [论文简析]SAC: Soft Actor-Critic Part 1[1801.01290] hat means estimation

[TOC] Title: Agent57: Outperforming the Atari Human Benchmark 2020 Author: Adria Badia et. al. Publish Year: 2020 Review Date: Nov 2021 Summary of paper This needs to be only 1-3 sentences, but it demonstrates that you understand the paper and, moreover, can summarize it more concisely than the author in his abstract. Agent57 is the SOTA Atari RL agent in 2020 that can play difficult Atari games like “Montezuma’s Revenge, “Pitfall”, “Solaris” and “Skiing”. ...

[TOC] Title: Width-based Lookaheads with Learnt Base Policies and Heuristics Over the Atari-2600 Benchmark Author: Stefan O’Toole et. al. Publish Year: 2021 Review Date: Tue 16 Nov 2021 Summary of paper This needs to be only 1-3 sentences, but it demonstrates that you understand the paper and, moreover, can summarize it more concisely than the author in his abstract. This paper proposed a new width-based planning and learning agent that can play Atari-2600 games (though it cannot play Montezuma’s Revenge). The author claimed that width-based planning exploration plus (greedy) optimal MDP policy exploitation is able to achieve better performance than Monte-Carlo Tree Search. ...