53 research outputs found
Locomotor activity: A distinctive index in morphine self-administration in rats.
No full textSelf-administration of addictive drugs is a widely used tool for studying behavioral, neurobiological, and genetic factors in addiction. However, how locomotor activity is affected during self-administration of addictive drugs has not been extensively studied. In our present study, we tested the locomotor activity levels during acquisition, extinction and reinstatement of morphine self-administration in rats. We found that compared with saline self-administration (SA), rats that trained with morphine SA had higher locomotor activity. Rats that successfully acquired SA also showed higher locomotor activity than rats that failed in acquiring SA. Moreover, locomotor activity was correlated with the number of drug infusions but not with the number of inactive pokes. We also tested the locomotor activity in the extinction and the morphine-primed reinstatement session. Interestingly, we found that in the first extinction session, although the number of active pokes did not change, the locomotor activity was significantly lower than in the last acquisition session, and this decrease can be maintained for at least six days. Finally, morphine priming enhanced the locomotor activity during the reinstatement test, regardless of if the active pokes were significantly increased or not. Our results clearly suggest that locomotor activity, which may reflect the pharmacological effects of morphine, is different from drug seeking behavior and is a distinctive index in drug self-administration
I See
No full textI See is a rescore of a cinematic piece that I selected to show my skills and the knowledge I have gained throughout my one-year study at Berklee College of Music. This composition represents a fusion of musical traditions, blending a full Western orchestra with traditional Chinese instruments to create a rich and unique piece. By incorporating the pentatonic scale and harmonies characteristic of Chinese music, I aimed to complement the visual narrative of the Chinese-style animation while authentically reflecting its cultural roots. Throughout this project, I carefully considered how musical elements such as theme development, orchestration, and harmonic language contribute to the emotional and narrative impact of the piece. The orchestration balances the textures of the Western orchestra with the distinct timbres of Chinese instruments, creating a dialogue between these two musical worlds. Additionally, the use of harmony and melodic material is designed to highlight the animation’s storytelling and enhance its cultural atmosphere. In this thesis, I will discuss my creative process in detail, focusing on the development of thematic material, the choices behind orchestration, the application of harmony, and how these musical components interact effectively with the visuals. This project reflects both my technical growth and artistic vision as a composer working at the intersection of Chinese and Western musical traditions.https://remix.berklee.edu/graduate-studies-scoring/1326/thumbnail.jp
Pyramid Graph Networks With Connection Attentions for Region-Based One-Shot Semantic Segmentation
No full textOne-shot image segmentation aims to undertake the segmentation task of a novel class with only one training image available. The difficulty lies in that image segmentation has structured data representations, which yields a many-to-many message passing problem. Previous methods often simplify it to a one-to-many problem by squeezing support data to a global descriptor. However, a mixed global representation drops the data structure and information of individual elements. In this paper, we propose to model structured segmentation data with graphs and apply attentive graph reasoning to propagate label information from support data to query data. The graph attention mechanism could establish the element-to-element correspondence across structured data by learning attention weights between connected graph nodes. To capture correspondence at different semantic levels, we further propose a pyramid-like structure that models different sizes of image regions as graph nodes and undertakes graph reasoning at different levels. Experiments on PASCAL VOC 2012 dataset demonstrate that our proposed network significantly outperforms the baseline method and leads to new state-of-the-art performance on 1-shot and 5-shot segmentation benchmarks.AI SingaporeMinistry of Education (MOE)Accepted versionThis work is supported by the National Research Foundation Singapore under its AI Singapore Programme [AISG-RP-2018-003] and the MOE Tier-1 research grant [RG126/17 (S)]. We would like to thank NVIDIA for GPU donation. Any opinions, findings and conclusions or recommendations expressed in this material are those of the author(s) and do not reflect the views of National Research Foundation, Singapore
Antinociceptive effects induced by intra-periaqueductal grey injection of the galanin receptor 1 agonist M617 in rats with morphine tolerance
No full textThe present study was performed to investigate the antinociceptive effects of M617, a selective galanin receptor 1 agonist, and M1145, a selective galanin receptor 2 agonist, in the periaqueductal grey (PAG) in rats with morphine tolerance. Intra-PAG injection of 0.1 nmol, 0.5 nmol and 1 nmol of M617 induced dose-dependent increases in hindpaw withdrawal latencies (HWLs) to noxious thermal and mechanical stimulations in rats with morphine tolerance. Nevertheless, intra-PAG injection of 5 nmol of the selective galanin receptor 2 agonist M1145 showed no significant influences on HWLs to noxious thermal and mechanical stimulations in rats with morphine tolerance. The results demonstrated that it is the selective galanin receptor 1 agonist M617, not the selective galanin receptor 2 agonist M1145, induced significant antinociceptive effects in morphine-tolerant rats, indicating that galanin receptor I is involved in nociceptive modulation in the PAG of morphine-tolerant rats. (C) 2013 Elsevier Ireland Ltd. All rights reserved.NeurosciencesSCI(E)PubMed2ARTICLE125-12855
Towards human-centered recommender systems
No full textRecently, there has been extensive interest in developing intelligent human-centered AI (artificial intelligence) systems that support human participation so as to facilitate cooperation between humans and machines. As one of the typical decision making paradigm in AI, recommender systems which have become an integral part of our lives, are a particularly pervasive form of AI system that can aid in decision-making in the face of ever-growing amounts of information. It now becomes imaginable and achievable with the help of advanced artificial intelligence, especially the modern deep learning based recommender systems that is known for its superior representation and predictive power, have made great strides in accuracy and effectiveness. Meanwhile, it also raises a number of important challenges: 1) How can we actively incorporate human participation into the decision-making procedure of recommender systems? It aims to integrate human participation as guidance to keep the decision-making process consistent with human feedback to maintain the trustworthiness to human beings. 2) How can we ensure that explanations are provided such that users can better understand why particular items are being recommended? In this aspect, explainable recommendation can be leveraged to not only assist the agent to provide high-quality recommendation results but also offering personalized and intuitive explanations with better user engagement, which are important for several modern recommender systems such as e-commerce and social media platforms etc. 3) How can we alleviate biases in recommender systems? Seldom progress has been explored to mitigate the biases that arise in human-centered recommender systems so as to hurt user satisfaction and trust towards the recommendation service. In this thesis, we proposes several novel methods to fill these gaps. In particular, for improved human understanding, we introduce an adversarial semantic learning framework for cross-lingual settings understanding. For human integration, a human-in-the-loop conversational recommender system with external graph structure is introduced. To ensure fair explanations, we mitigate the unfairness within graph-based explainable reasoning in the recommender system. Finally, for human-system cooperation, we present a popularity debiasing framework to integrate user interaction and debiased dialogue stat management in a conversational recommender system. We not only extensively evaluate our proposed approaches on multiple real-world recommendation datasets, but also contribute open public datasets to the community. The experimental results demonstrate the effectiveness of the proposed methods in achieving satisfying prediction accuracy, mitigating bias, and providing users with understandable explanations.Ph.D.Includes bibliographical reference
Neural logic reasoning and applications
No full textRecent years have witnessed the success of deep neural networks in many research areas. The fundamental idea behind the design of most neural networks is to learn similarity patterns from data for prediction and inference, which lacks the ability of cognitive reasoning. However, the concrete ability of reasoning is critical to many theoretical and practical problems. On the other hand, traditional symbolic reasoning methods do well in making logical inferences, but they are mostly hard rule-based reasoning, which limits their generalization ability to different tasks since different tasks may require different rules. Both reasoning and generalization ability are important for prediction tasks such as recommender systems, where reasoning provides a strong connection between user history and target items for accurate prediction, and generalization helps the model to draw a robust user portrait over noisy inputs. In this dissertation, we propose a Neural Logical Reasoning (NLR) framework to integrate the power of deep learning and logical reasoning. NLR is a dynamic modularized neural architecture that learns basic logical operations such as AND, OR, NOT as neural modules, and conducts propositional logical reasoning through the logical structured network for inference. In order to examine the effectiveness of the neural logical reasoning concept, we do extensive studies including (i) Conducting experiments on theoretical tasks to verify that our proposed framework has the ability to conduct logical inference; (ii) Applying the neural logical reasoning framework to recommendation task to explore its ability of solving real-word problems; (iii) Extending the application to more general graph-related tasks such as knowledge graph completion. Experiments show that our approach achieves state-of-the-art performance in various application scenarios. Moreover, we utilize the neural architecture search strategy to allow the model to learn the adaptive logical neural architectures automatically which brings flexibility to our framework.Ph.D.Includes bibliographical reference
Multi-dimensional federated learning in recommender systems
No full textA wide range of web services like e-commerce, job-searching, and target advertising heavily rely on recommender systems that find products of interest to fulfill users' diverse and complicated demands. To better model the user preferences and provide satisfactory recommendations, there has been an increasing research focus on constructing more accurate and complete user representations that exploit the user's personal information including the profile and the behavior history. Inevitably, this would induce privacy risks for users in two main aspects: 1) collection of users' sensitive attributes like gender and address; 2) untrustworthy exchange of user data among services. Thus, this natural conflict between user privacy and recommendation accuracy has drawn lots of attention in recent years. One of the most widely studied and verified solution is the federated learning technique. The general idea behind federated learning is to maintain users' critical data on edge devices (e.g. mobile phones) and communicate only the model parameters to the central server.
However, a standard federated learning system is designed for a single task with a simple learning objective. In reality, a user typically interacts with various applications every day with heterogeneous intentions. In this dissertation, I will extend the federated learning system to this realistic but more complex multi-objective setting, where multiple federated learning agents collaborate in an environment with multiple central servers and a number of distributed edge devices: a) the set of centralized services possess the collective knowledge of all users but each service collect its own domain data, and b) the set of personalized smart assistants on every user's edge devices maintain the complete data of users and help provide satisfactory personal experiences through interactions with different services. The resulting framework, as a multi-dimensional federated learning paradigm, instantiates the direct intersection between the two trends: demands for a more complete user profile and more protection of user privacy. In the foreseeable future, this problem would become a fundamental piece of the omnipotent personal intelligent assistant.
Formally, this problem is a multi-task federated optimization problem, and I identify several main challenges in this work: First, task dimensions are heterogeneous not only in their inherent design of user profiles and learning objectives but also heterogeneous in terms of the users involved in the learning process; Second, the cross-dimension nature induces requirements of extra privacy control among services, in addition to each task's privacy protection of users. Third, the global objective of one service may violate user privacy or conflict with those of other services even in the federated environment. Fourth, collaboration among multiple distributed learning systems may induce extra communication and computation overhead, which makes the entire system unstable and slow. I will present the general framework of multi-dimensional federated learning, and illustrate several solution techniques including privacy-based information management, indirect transfer, and objective transformation that address the aforementioned challenges.Ph.D.Includes bibliographical reference
Target Detection Algorithm Incorporating Visual Expansion Mechanism and Path Syndication
No full textBecause the lack of semantic information exchange between characteristic layers, the SSD (Single Shot multibox Detector) algorithm has insufficient detection performance. To address this problem, a detection algorithm called VPE-SSD (Visual Path Enhancement SSD) by incorporating a visual expansion mechanism and path syndication proposed in this paper. Firstly, a visual expansion mechanism is added to the shallow characteristic layer to increase the perceptual field. This enables the semantic information in the shallow layer to be more fully utilized by the network. It can also achieve the purpose of enhancing the expressiveness of the shallow feature layer. Then, the processed deep and shallow characteristic layers are fed into the path syndication module for bi-directional fusion. This improves the global information of the feature layers and generates multi-scale global feature maps. Next, to enhance the detailed information of deep characteristics and improve their expression, the deep characteristic enhancement module is applied to the last three characteristic maps. Finally, using the blended attention module to reduce the negative interference and improve the expression of characteristic maps during target detection. The experimental analysis of the VPE-SSD algorithm is conducted on VOC and COCO, and the mAP is 83.4% and 48.4%. From the result, VPE-SSD algorithm can make better use of the different size characteristic information which helps to improve the performance of the algorithm
- …
