The document discusses the benefits of exercise for mental health. Regular physical activity can help reduce anxiety and depression and improve mood and cognitive functioning. Exercise causes chemical changes in the brain that may help protect against mental illness and improve symptoms.
http://imatge-upc.github.io/telecombcn-2016-dlcv/
Deep learning technologies are at the core of the current revolution in artificial intelligence for multimedia data analysis. The convergence of big annotated data and affordable GPU hardware has allowed the training of neural networks for data analysis tasks which had been addressed until now with hand-crafted features. Architectures such as convolutional neural networks, recurrent neural networks and Q-nets for reinforcement learning have shaped a brand new scenario in signal processing. This course will cover the basic principles and applications of deep learning to computer vision problems, such as image classification, object detection or text captioning.
Slides contain basics of python programming: Pattern Matching, Regular expression library, Basic RE Characters, RE methods, Programming examples and Demonstrations.
Glove global vectors for word representationhyunyoung Lee
The document discusses the benefits of exercise for mental health. Regular physical activity can help reduce anxiety and depression and improve mood and cognitive function. Exercise causes chemical changes in the brain that may help protect against mental illness and improve symptoms for those who already suffer from conditions like depression and anxiety.
The document discusses the benefits of exercise for both physical and mental health. It notes that regular exercise can reduce the risk of diseases like heart disease and diabetes, improve mood, and reduce feelings of stress and anxiety. The document recommends that adults get at least 150 minutes of moderate exercise or 75 minutes of vigorous exercise per week to gain these benefits.
Weakly Supervised Learning: Introduction and Best Practices
In the talk we will introduce the definition of three main types of weakly supervised learning: incomplete, inexact and inaccurate; we examine how the models can be trained in case of weak supervision and view the real application of weakly supervised learning, how it can improve results and decrease the costs.
Bio:
Kristina Khvatova works as a Software Engineer at Softec S.p.A. Currently she is involved in the development of a project for data analysis and visualisation; it includes quantitative and qualitative analysis based on classification, optimisation, time series prediction, anomaly detection techniques. She obtained a master degree in Mathematics at the Saint-Petersburg State University and a master degree in Computer Science at the University of Milano-Bicocca.
The document discusses the benefits of exercise for mental health. Regular physical activity can help reduce anxiety and depression and improve mood and cognitive function. Exercise causes chemical changes in the brain that may help protect against mental illness and improve symptoms for those who already suffer from conditions like anxiety and depression.
1) Deep learning is a type of machine learning that uses neural networks with many layers to learn representations of data with multiple levels of abstraction.
2) Deep learning techniques include unsupervised pretrained networks, convolutional neural networks, recurrent neural networks, and recursive neural networks.
3) The advantages of deep learning include automatic feature extraction from raw data with minimal human effort, and surpassing conventional machine learning algorithms in accuracy across many data types.
The document discusses the benefits of exercise for mental health. Regular physical activity can help reduce anxiety and depression and improve mood and cognitive functioning. Exercise causes chemical changes in the brain that may help protect against mental illness and improve symptoms.
http://imatge-upc.github.io/telecombcn-2016-dlcv/
Deep learning technologies are at the core of the current revolution in artificial intelligence for multimedia data analysis. The convergence of big annotated data and affordable GPU hardware has allowed the training of neural networks for data analysis tasks which had been addressed until now with hand-crafted features. Architectures such as convolutional neural networks, recurrent neural networks and Q-nets for reinforcement learning have shaped a brand new scenario in signal processing. This course will cover the basic principles and applications of deep learning to computer vision problems, such as image classification, object detection or text captioning.
Slides contain basics of python programming: Pattern Matching, Regular expression library, Basic RE Characters, RE methods, Programming examples and Demonstrations.
Glove global vectors for word representationhyunyoung Lee
The document discusses the benefits of exercise for mental health. Regular physical activity can help reduce anxiety and depression and improve mood and cognitive function. Exercise causes chemical changes in the brain that may help protect against mental illness and improve symptoms for those who already suffer from conditions like depression and anxiety.
The document discusses the benefits of exercise for both physical and mental health. It notes that regular exercise can reduce the risk of diseases like heart disease and diabetes, improve mood, and reduce feelings of stress and anxiety. The document recommends that adults get at least 150 minutes of moderate exercise or 75 minutes of vigorous exercise per week to gain these benefits.
Weakly Supervised Learning: Introduction and Best Practices
In the talk we will introduce the definition of three main types of weakly supervised learning: incomplete, inexact and inaccurate; we examine how the models can be trained in case of weak supervision and view the real application of weakly supervised learning, how it can improve results and decrease the costs.
Bio:
Kristina Khvatova works as a Software Engineer at Softec S.p.A. Currently she is involved in the development of a project for data analysis and visualisation; it includes quantitative and qualitative analysis based on classification, optimisation, time series prediction, anomaly detection techniques. She obtained a master degree in Mathematics at the Saint-Petersburg State University and a master degree in Computer Science at the University of Milano-Bicocca.
The document discusses the benefits of exercise for mental health. Regular physical activity can help reduce anxiety and depression and improve mood and cognitive function. Exercise causes chemical changes in the brain that may help protect against mental illness and improve symptoms for those who already suffer from conditions like anxiety and depression.
1) Deep learning is a type of machine learning that uses neural networks with many layers to learn representations of data with multiple levels of abstraction.
2) Deep learning techniques include unsupervised pretrained networks, convolutional neural networks, recurrent neural networks, and recursive neural networks.
3) The advantages of deep learning include automatic feature extraction from raw data with minimal human effort, and surpassing conventional machine learning algorithms in accuracy across many data types.
Acceleration of Deep Neural Networks Using Stochastic Computing (확률컴퓨팅을 이용한 딥...NAVER Engineering
발표자: 심현욱(UNIST 박사과정)
발표일: 2018.3.
최근 들어 다시 각광받는 인공지능은 deep neural network의 (딥뉴럴넷) 발전이 크게 뒷받침하였다고 할 수 있습니다. 여러 스타트업에서도 다양한 아이디어를 가지고 저전력 IoT 디바이스 등에 적용하고자 하기도 하지만, 이 인공신경망을 실행하는 데에 따르는 많은 연산량이 난제로 남아있습니다. Stochastic computing(확률컴퓨팅)은 기존 바이너리 컴퓨팅과 다른 패러다임으로, 저전력을 장점으로 딥뉴럴넷 가속에 대한 대안으로 연구되어 왔습니다. 본 발표에서는 확률컴퓨팅으로 딥뉴럴넷을 가속하는 연구와 그 한계, 그리고 미래를 이야기하고자 합니다.
Super tickets in pre trained language modelsHyunKyu Jeon
This document discusses finding "super tickets" in pre-trained language models through pruning attention heads and feedforward layers. It shows that lightly pruning BERT models can improve generalization without degrading accuracy (phase transition phenomenon). The authors propose a new pruning approach for multi-task fine-tuning of language models called "ticket sharing" where pruned weights are shared across tasks. Experiments on GLUE benchmarks show their proposed super ticket and ticket sharing methods consistently outperform unpruned baselines, with more significant gains on smaller tasks. Analysis indicates pruning reduces model variance and some tasks share more task-specific knowledge than others.
Synthesizer rethinking self-attention for transformer models HyunKyu Jeon
The document expresses gratitude to the reader for taking the time to listen. It does not provide any other details, context, or information beyond thanking the reader for listening. The summary captures the essence of the document in a single concise sentence.
This document summarizes Meta Back-Translation, a method for improving back-translation by training the backward model to directly optimize the performance of the forward model during training. The key points are:
1. Back-translation typically relies on a fixed backward model, which can lead the forward model to overfit to its outputs. Meta back-translation instead continually trains the backward model to generate pseudo-parallel data that improves the forward model.
2. Experiments show Meta back-translation generates translations with fewer pathological outputs like greatly differing in length from references. It also avoids both overfitting and underfitting of the forward model by flexibly controlling the diversity of pseudo-parallel data.
3. Related work leverages mon
Maxmin qlearning controlling the estimation bias of qlearningHyunKyu Jeon
This document summarizes the Maxmin Q-learning paper published at ICLR 2020. Maxmin Q-learning aims to address the overestimation bias of Q-learning and underestimation bias of Double Q-learning by maintaining multiple Q-functions and using the minimum value across them for the target in the Q-learning update. It defines the action selection and target construction for the update based on taking the maximum over the minimum Q-value for each action. The algorithm initializes multiple Q-functions, selects a random subset to update using the maxmin target constructed from the minimum Q-values. This approach reduces the biases seen in prior methods.
3. 10-MIN
DEEP-LEARNING INTRODUCTION
인공지능? 머신러닝? 딥러닝 ?
인공지능(Artificial Intelligence): 특정 분야를 지칭하는 것이 아닌, 지능적 요소가 포함된 기술을 총칭
머신러닝(Machine Learning): ‘데이터’에서 ‘모델’을 스스로 찾아내는 기법
딥러닝(Deep Learning)
:심층 신경망을 이용한 머신러닝 기법
참고: www. euclidean.com
4. 10-MIN
DEEP-LEARNING INTRODUCTION
머신러닝의 분류
지도학습(Supervised Learning)
비지도학습(Unsupervised Learning)
강화학습(Reinforcement Learning)
데이터와 그에 대응되는 예측결과 값(Label)을 투입하여 서로 간의 관계를 학습하고, 해당 데이터와 일치
또는 유사한 데이터가 입력되었을때, 학습시킨 관계에 따른 결과 값을 내도록 하는 것.
결과 값이 없는 데이터들을 입력하여 각각의 데이터들에 내재된 속성을 기반으로 분류 등의 학습을 하고, 새로운
데이터가 입력되었을때 해당 데이터의 내제된 속성에 따라 학습된 결과를 도출하는 것.
에이전트(agent)가 특정 상태에 대한 반응으로서의 행동(action)를 내보내면, 이에 따른 보상(reward) 또는
벌칙(penalty)을 주어 달성하고자 하는 목표 결과(action)을 내보내도록 학습하는 것.
SUPERVISED LEARNING UNSUPERVISED LEARNING REINFORCEMENT LEARNING
Data
OUTPUT
INPUT
Label
MODEL
입력된 레이블(Label)과
결과값(Output)의 차이를 최소화
Data
OUTPUT
INPUT
MODEL
데이터의 내제된 특성(Feature)의
유사성, 관련성을 바탕으로 학습
Agent
Environment
ACTION
STATE / REWARD
에이전트와 환경 간의 주고 받음(action//state, reward)
을 통해 조성[Shaping]하는 것
5. 10-MIN
DEEP-LEARNING INTRODUCTION
신경망; 지식은 흐른다Neural Network
Dendrite
Cell body
Node of Ranvier
Axon Terminal
Schwann cell
Myelin sheath
Axon
Nucleus
- https://commons.wikimedia.org/wiki/File:Neuron.svg
자극이 주어짐
(역치 이상) 밖의 나트륨 흡수
나트륨이 말단까지 흡수/확산
말단에서 화학물질 생성
다른 뉴런의 나트륨 흡수를 도움
결국, 신경은 나트륨의 확산일 뿐이고
인지 및 기억은 각 뉴런들의 연결 형태
수상 돌기
축삭 돌기
미엘린 수초
시반 세포
신경 세포체
핵
랑비에 결절
축색 종말
6. 10-MIN
DEEP-LEARNING INTRODUCTION
신경망; 지식은 흐른다Neural Network
Stimulus
Na+
Na+
Na+
Na+
Na+
전자의 위치에너지 차이(전위 차)에 의해 나트륨 이온이
이동하여 시냅스에서의 화학물질 전달을 촉진[신호전달]
시냅스이온에 의한 전기적 신호가 강할수록
시냅스에서의 화학적 작용이 활성화 Synapse
Na+
Na+
Na+
Na+ Na+
Na+
Na+
Na+
Na+
Na+
Na+
Na+
Na+
Na+ Na+
Na+
Na+
Na+
Na+
Na+
Na+
Na+
Na+
Na+
Na+
Na+
Na
Na+ Na+
Na+
Na+
Na+
Na+
뉴런 내부
Inside of Neuron
자극
7. 10-MIN
DEEP-LEARNING INTRODUCTION
신경망과 인공신경망
Neural Network and Artificial Neural Network
자극이 주어짐신경망 일정 수준 이상 (역치)에 도달 화학물질을 통해 다음 뉴런을 활성화
데이터가 주어짐인공신경망 가중치를 곱함 / 활성화 함수 다음 계층에게 연산된 데이터를 전달
WHY? 1. 이전 뉴런에서 신호를 받아도 시냅스 구조가 달라 활성화 정도의 차이가 발생.
2. 전(全) 연결구조를 이용한 행렬 연산을 통해 효율적인 연산 수행 가능.