Mikayel Samvelyan, Developer in London, United Kingdom
Mikayel is available for hire
Hire Mikayel

Mikayel Samvelyan

Verified Expert  in Engineering

Machine Learning Developer

Location
London, United Kingdom
Toptal Member Since
May 1, 2018

With a master's degree from the University of Oxford, Mikayel是一名机器学习专家和数据科学家，专门研究自然语言处理, computer vision, and reinforcement learning. 他在深度学习解决方案的研究和开发方面拥有丰富的经验，并曾在Reddit等公司工作, Mentor, and USC. Mikayel在领先的机器学习会议上与人合作撰写了几篇科学出版物.

Deep LearningMachine LearningNatural Language Processing (NLP)PythonC++PyTorchTensorFlowRoboticsData VisualizationUnixJupyterSciPyNumPyPandasFlask

Availability

Part-time

Preferred Environment

PyCharm, Vi

The most amazing...

...我所从事的项目是开发一种新的AI算法来掌握《欧博体育app下载》的游戏.

Work Experience

Research Engineer (Computer Vision)

2020 - 2020
USC Information Sciences Institute (via Toptal)
  • 为视频预测模型构建了一个可扩展的管道，该管道在多个远程GPU实例上并行运行.
  • 执行姿态估计，并将3D对象投影到视频中相应的2D投影.
  • 使用Python和OpenCV开发了一个运动检测软件.
  • 使用OpenCV和Dlib构建了一个头部姿态估计库.
  • Developed a real-time object detection library based on YOLO.
  • 在USC ISI的现场访问期间，该工作已提交给DARPA.
Technologies: OpenCV, Python

Machine Learning Expert

2019 - 2019
Reddit (via Toptal)
  • 使用PyTorch为大规模自然语言处理任务开发深度学习解决方案.
  • 微调预训练的NLP模型，如BERT、XLNet和RoBERTa.
  • 使用LDA、NMF等方法进行主题建模实验.
  • 为大规模自然语言分类任务构建模块化管道.
  • Analyzed large-scale datasets using BigQuery.
Technologies: Amazon SageMaker, Custom BERT, PyTorch, Python

Data Scientist

2018 - 2019
Highlander Technology, Inc. (via Toptal)
  • 使用PyTorch为大规模自然语言处理任务开发和优化深度学习解决方案.
  • 实现了对大量数据的噪声和新颖性检测.
  • Performed rigorous data mining for raw and noisy data.
  • 创建了一个API端点，将模型用于多类分类.
  • 构建了大规模多类分类的机器学习流水线.
Technologies: Amazon SageMaker, PyTorch, Python

Machine Learning Consultant

2018 - 2018
Mission Ready Marketing, LLC (via Toptal)
  • 为推荐引擎中使用的机器学习方法设计了一个管道.
  • 研究了在没有历史数据的情况下使用无监督学习技术构建推荐系统的方法.
  • 分析了基于物品、基于用户、基于矩阵分解和混合推荐系统.
  • 创建了完整的技术方法文档，用于将项目的学习组件与数据库和后端连接起来.
Technologies: Python

Machine-learning Researcher

2018 - 2018
XIX.ai
  • 在卫星、无人机和地面图像上搜索和识别物体.
  • 设计并实现了一种大规模车队管理的深度强化学习算法.
  • Created a simulator for a ride-hailing service using Python.
  • 建立了一个机器学习框架来预测用户的意图.
Technologies: Google Vision API, Sacred, PyTorch, Python

Research and Development Engineer

2014 - 2016
Mentor Graphics
  • 对逻辑优化、划分、放置和路由的算法进行了研究.
  • Developed a unified environment for design capturing, simulation setup, verification, and analysis for a custom integrated circuit design platform.
  • Integrated the environment with third-party IDE software.
  • 使用Lex和Yacc创建了各种可重用的编译器.
Technologies: Amazon Lex, Yacc, C++

Software Engineer

2012 - 2013
Instigate Robotics
  • Built firmware and software for robotic applications.
  • Developed 3D printing technologies.
  • 用c++和Qt创建应用程序的图形用户界面(GUI).
  • 在STM32和Arduino mcu上设计和实现各种嵌入式应用.
  • Built an educational development environment for robotics.
Technologies: Qt, Smalltalk, C++, C

Software Engineering Intern

2012 - 2012
Instigate Design
  • 在多处理器计算机上为并行计算开发了一个独立于硬件/软件的环境.
  • Supported the development of a compiler's front end.
Technologies: Clang, C++

深度多智能体强化学习的单调值函数分解

http://proceedings.mlr.press/v80/rashid18a.html
QMIX是一种最先进的基于值的协同深度多智能体强化学习算法，用于集中训练和分散执行的设置.
QMIX采用神经网络，将联合动作值估计为每个代理值的复杂非线性组合，仅以局部观察为条件. 我们从结构上强制联合作用值在每个代理值中是单调的, 在非策略学习中，哪一种方法可以实现联合行动价值的可处理最大化, 并保证集中和分散政策之间的一致性.

QMIX在ICML 2018上发布，ICML 2018是机器学习研究的主要会议之一.

PyMARL: Python Multi-agent Reinforcement Learning

http://github.com/oxwhirl/pymarl
PyMARL是一个用于深度多智能体强化学习的研究平台，它允许开箱即用的实验和开发.

Written in PyTorch, PyMARL提供了一些最先进的方法的实现，比如QMIX, COMA, and independent Q-Learning.

In collaboration with the Berkeley AI Research lab, 以上的一些算法也已经成功移植到可扩展的RLlib框架中.

SMAC: The StarCraft Multi-agent Challenge

http://github.com/oxwhirl/smac
星际争霸多智能体挑战(SMAC)是合作多智能体强化学习(MARL)的一个基准，它提供了部分可观察性的元素, challenging dynamics, and high-dimensional observation spaces. SMAC is built using the StarCraft II game engine, 为协作式强化学习的研究创造了一个测试平台，其中每个游戏单元都是一个独立的强化学习代理.

Paper: http://arxiv.org/abs/1902.04043
Blogpost: http://whirl.cs.ox.ac.uk/blog/smac/
Blogpost by NVIDIA: http://nvda.ws/2I88F1I

SMAC was presented at AAMAS 2019, agent和多agent系统领域最大和最具影响力的会议.

ADAM Visual Perception

http://github.com/isi-vista/adam-visual-perception
我们的学习过程依赖于拥有一个相当丰富的(尽管在发展上是合理的)输入表征. 本知识库探讨了如何通过算法捕获视觉感知的两个方面，这两个方面对早期语言学习至关重要, namely motion cause detection and gaze object detection.

Languages

C++, C#, Python, Java, Lua, SQL, R, C, Objective-C, Smalltalk, Assembler x86, Prolog, Lisp, JavaScript, Octave

Frameworks

Flask, .NET, Qt, Spark

Libraries/APIs

SciPy, NumPy, Pandas, PyTorch, TensorFlow, Scikit-learn, Keras, OpenCV, Spark ML, PySpark, Google Vision API, Google APIs

Tools

Jupyter, LaTeX, Makefile, Vim Text Editor, Subversion (SVN), Git, PyCharm, CVS, MATLAB, Tmux, Tableau, Amazon Lex, Sacred, BigQuery, Amazon SageMaker, AutoML

Platforms

Unix、Docker、MacOS、Linux、Amazon EC2、Amazon Web Services (AWS)

Other

System Programming, Deep Learning, Reinforcement Learning, Natural Language Processing (NLP), Custom BERT, Statistics, Classification, Predictive Modeling, Machine Learning, Mathematics, Data Visualization, GPT, Generative Pre-trained Transformers (GPT), Robotics, Computer Vision, Recommendation Systems, Big Data, Image Processing, Physics, Vi, Yacc, Clang, Time Series Analysis

Paradigms

Requirements Analysis, Design Patterns, Data Science

Storage

MongoDB, PostgreSQL, Redshift

2016 - 2017

Master of Science Degree in Computer Science

University of Oxford - Oxford, UK

2014 - 2016

计算机科学与应用数学硕士学位

Yerevan State University - Yerevan, Armenia

2010 - 2014

计算机科学与应用数学学士学位

Yerevan State University - Yerevan, Armenia

2013 - 2013

Undergraduate Exchange in Computer Science

Delta State University - Cleveland, MS, USA

AUGUST 2015 - PRESENT

CS190.1x: Scalable Machine Learning

The University of California, Berkeley via edX

MAY 2015 - PRESENT

Algorithms: Design and Analysis, Part 2

Stanford University via Coursera

APRIL 2015 - PRESENT

Algorithms: Design and Analysis, Part 1

Stanford University via Coursera