Christopher Blake

Christopher Blake

Oklahoma City, Oklahoma, United States
845 followers 500+ connections

Activity

Join now to see all activity

Experience

  • GitHub Graphic

    GitHub

    Oklahoma City, Oklahoma, United States

  • -

    Oklahoma City, Oklahoma

  • -

    Hanover, Germany

  • -

    Celle, Germany

  • -

    The Woodlands, Texas

  • -

    West Lafayette, IN

  • -

    West Lafayette, Indiana

  • -

    West Lafayette, Indiana

  • -

    Cicero, Indiana

  • -

    Marion, Indiana

Education

Licenses & Certifications

Volunteer Experience

  • Junior Achievement of Southeast Texas Graphic

    Volunteer Teacher

    Junior Achievement of Southeast Texas

    - 2 years 4 months

    Children

    Visit a few classrooms of children every quarter and teach real-life principles. The goal is to be a source of inspiration, someone to look-up up to, and to make their day a little less routine (aka more fun).

  • IndyFringe Graphic

    Windows System Administrator

    IndyFringe

    - 5 months

    Arts and Culture

Publications

  • Reinforcement Learning Based Decision Tree Induction Over Data Streams with Concept Drifts

    IEEE

    Traditional decision tree induction algorithms are greedy with locally-optimal decisions made at each node based on splitting criteria like information gain or Gini index. A reinforcement learning approach to decision tree building seems more suitable as it aims at maximizing the long-term return rather than optimizing a short-term goal. In this paper, a reinforcement learning approach is used to train a Markov Decision Process (MDP), which enables the creation of a short and highly accurate…

    Traditional decision tree induction algorithms are greedy with locally-optimal decisions made at each node based on splitting criteria like information gain or Gini index. A reinforcement learning approach to decision tree building seems more suitable as it aims at maximizing the long-term return rather than optimizing a short-term goal. In this paper, a reinforcement learning approach is used to train a Markov Decision Process (MDP), which enables the creation of a short and highly accurate decision tree. Moreover, the use of reinforcement learning naturally enables additional functionality such as learning under concept drifts, feature importance weighting, inclusion of new features and forgetting of obsolete ones as well as classification with incomplete data. To deal with concept drifts, a reset operation is proposed that allows for local re-learning of outdated parts of the tree. Preliminary experiments show that such an approach allows for better adaptation to concept drifts and changing feature spaces, while still producing a short and highly accurate decision tree.

    See publication
  • Usage of Finite Element Analysis to Produce Trends of 'Build Rate' vs 'Weight on Bit' to Improve Well Planning, while-Drilling Operations, Tool Design, and Post Job Analysis

    Downhole motors are used for directional drilling. Many design parameters such as AKO angle and stabilization type have documentation to show their effect on the build rate. However, weight on bit and its effect on the downhole geometry is often not considered. A 'Build Rate' vs 'Weight on Bit' chart, based on FEA calculations with different motor configurations, is proposed to demonstrate these effects. By utilizing such a chart, significant improvements are possible to tool design, wellbore…

    Downhole motors are used for directional drilling. Many design parameters such as AKO angle and stabilization type have documentation to show their effect on the build rate. However, weight on bit and its effect on the downhole geometry is often not considered. A 'Build Rate' vs 'Weight on Bit' chart, based on FEA calculations with different motor configurations, is proposed to demonstrate these effects. By utilizing such a chart, significant improvements are possible to tool design, wellbore planning, and in field operational corrections.

    See publication

Patents

  • Hydraulic tools including removable coatings, drilling systems, and methods of making and using hydraulic tools

    Issued US US9896885B2

    A hydraulic tool includes a stator, a rotor, and a removable coating. At least one of the stator and the rotor comprises a resilient material. The removable coating has a thickness selected to compensate for expected swelling of the resilient material or an expected contraction of a clearance between the rotor and the stator based on thermal expansion. The removable coating is disposed on a surface of at least one of the rotor and the stator, and the removable coating is formulated to be…

    A hydraulic tool includes a stator, a rotor, and a removable coating. At least one of the stator and the rotor comprises a resilient material. The removable coating has a thickness selected to compensate for expected swelling of the resilient material or an expected contraction of a clearance between the rotor and the stator based on thermal expansion. The removable coating is disposed on a surface of at least one of the rotor and the stator, and the removable coating is formulated to be removed during operation of the hydraulic tool. A method of operating a hydraulic tool includes passing a fluid through the hydraulic tool during rotation of the rotor within the stator and removing at least a portion of the removable coating responsive to rotation of the rotor within the stator as a volume of the resilient material increases responsive to contact with the fluid passing through the hydraulic fluid.

    See patent

Courses

  • Cognitive and Multiagent Systems

    -

  • Computing Systems

    -

  • Electrical Drives

    -

  • Engineering Dynamics and Vibration

    -

  • Forming Technology, Aspects of Process Design

    -

  • History and Methodology of Informatics and Computer Science

    -

  • Intelligent Computing

    -

  • Intelligent Systems

    -

  • Knowledge Management and Knowledge Engineering

    -

  • Mathematical Modelling and Simulation

    -

  • Methods of Optimization

    -

  • Micro- and Nanosystems

    -

  • Modern Problems of Computers and Information Science

    -

  • Neuroinformatics and Neurotechnologies

    -

  • Production of Optoelectronic Systems

    -

  • Software Development Technology

    -

Projects

  • Journal: Reinforcement Learning Based Decision Tree Induction over Data Streams with Concept Drifts

    -

    A journal extension of the similarily-named publication in 2018.
    Estimated completion: Early July, 2019.

  • Master Thesis: Knowledge Production and Control of a Black Box Using Machine Learning

    -

    An adaptive stream-based process is described for learning a set of vocabulary to control a black box. The method serves to accurately discretize the value space, enable tracking, and identify simple patterns. These values and patterns become the system knowledge (or vocabulary) and are used for training a reinforcement learning based decision tree. An interpretation layer enables developing higher level knowledge with time, creating an easy-to-read policy of the black box functionality…

    An adaptive stream-based process is described for learning a set of vocabulary to control a black box. The method serves to accurately discretize the value space, enable tracking, and identify simple patterns. These values and patterns become the system knowledge (or vocabulary) and are used for training a reinforcement learning based decision tree. An interpretation layer enables developing higher level knowledge with time, creating an easy-to-read policy of the black box functionality, providing control information. Tests are performed to demonstrate the effectiveness and limitations with open-loop systems. Finally, a proposal is made for further knowledge identification, closed-loop systems, and learning with less prior information.

    See project
  • SPbPU Online Control Portal

    -

    A web based solution is developed for configuring and control of a SCADA based manufacturing process. Three components are developed: a local control software for the SCADA system, a cloud portal for storage and semi-realtime synchronization, and a mobile-friendly web interface for system configuration and displaying the live system.

    See project
  • BHA Sys Pro Iterator + Toolkit + Web Interface

    -

    An internal-only tool for Baker Hughes Inc. for fully automated directional drilling prediction with downhole motors under various load conditions and design permutations.

  • PDC Test Drive

    -

    An internal-only tool for Baker Hughes Inc. for fully automated analysis of diamond bit response under several usage conditions.

  • Downhole Motors - Adjustable Bend Feasibility

    -

    An internal-only report for Baker Hughes Inc. that investigated the feasibility and usage details of an adjustable bend motor.

  • Downhole Motors - Design Considerations

    -

    An internal-only report for Baker Hughes Inc. that investigated the impact of several design factors for downhole motors and their influence on stresses and directional drilling performance.

  • Manipulation of Cuttings Generation

    -

    An internal-only report for Baker Hughes Inc. that investigated various techniques for affecting the generation of cuttings on a diamond bit.

  • Sidewalk Chalk Printer

    -

    In 2007 fliering on Purdue campus sidewalks was banned. Since then, the degree of passive marketing for its students has significantly diminished, leaving tedious hand chalking as the only effective method. To bring back effective fliering, the Purdue Sidewalk Printing Team consisting of 4 mechanical engineers and 2 industrial designers has developed an aesthetically pleasing and environmentally friendly device that will transfer standard .jpg and .gif images of flyers onto sidewalks using…

    In 2007 fliering on Purdue campus sidewalks was banned. Since then, the degree of passive marketing for its students has significantly diminished, leaving tedious hand chalking as the only effective method. To bring back effective fliering, the Purdue Sidewalk Printing Team consisting of 4 mechanical engineers and 2 industrial designers has developed an aesthetically pleasing and environmentally friendly device that will transfer standard .jpg and .gif images of flyers onto sidewalks using liquid chalk. The user need only plug a USB drive into the handle, choose upload from the visual interface, and select a flyer to begin printing.

    See project

Honors & Awards

  • Best Student Paper Award

    IEEE

  • Interview, PZH Magazine

    Produktionstechnisches Zentrum Hannover

    1 of 2 people selected by PZH Magazine to share insight about the multicultural and internal parts of the master's degree program at Leibniz University.

  • Strategic Partnership Program Scholarship

    Leibniz University

    A program designed for supporting international collaboration between Leibniz University and Universities in other countries.

  • Core Values and Recognition Award

    Baker Hughes Inc.

  • Core Values and Recognition Award

    Baker Hughes Inc.

  • Technology Excellence Award

    Baker Hughes

    Awarded for creating a verification system on bit orders, which saved the company several thousand dollars per month.

  • Senior Design Competition (2nd Place)

    Purdue University - Mechanical Engineering

    Received 2nd Place - Sidewalk Chalk Printer

Languages

  • English

    Native or bilingual proficiency

  • German

    Professional working proficiency

  • Russian

    Elementary proficiency

Organizations

  • Ecstatic Dance Houston

    -

    -
  • Suchu Dance Company

    Performer

    -
  • Purdue Contemporary Dance Company

    Officer

    -

Recommendations received

More activity by Christopher

View Christopher’s full profile

  • See who you know in common
  • Get introduced
  • Contact Christopher directly
Join to view full profile

Other similar profiles

Explore top content on LinkedIn

Find curated posts and insights for relevant topics all in one place.

View top content

Add new skills with these courses