BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20210402T160553Z
LOCATION:
DTSTART;TZID=America/New_York:20201109T090000
DTEND;TZID=America/New_York:20201109T200000
UID:submissions.supercomputing.org_SC20_sess280_job154@linklings.com
SUMMARY:Adversarial Machine Learning Postdoctoral Scholar
DESCRIPTION:Job Posting\n\nAdversarial Machine Learning Postdoctoral Schol
 ar\n\n\n\nAdversarial Machine Learning Postdoctoral Scholar - 91189\nOrgan
 ization: CR-Computational Research\n\nThe Data Science and Technology (DST
 , https://dst.lbl.gov/) department in the Computational Research Division 
 (https://crd.lbl.gov/) has an immediate opening for a post-doctoral resear
 cher to perform research and development in adversarial machine learning m
 ethods for complex control systems driven by reinforcement learning.\n\nTh
 e goal of this position is to research adversarial machine learning method
 s that will enable safer operation of automated, adaptive deep-learning-dr
 iven cyber-physical system processes. Numerous DOE-relevant processes are 
 becoming automated and adaptive, using machine learning techniques, such a
 s reinforcement learning, in which the state after one run of the process 
 automatically influences the actions taken on subsequent runs often withou
 t a human in the loop. Examples of such processes relevant to DOE today in
 clude: intelligent transportation systems, adaptive control of grid-attach
 ed equipment to stabilize power grid function, and more. This creates a vu
 lnerability for a cyber attacker to sabotage processes through tainted tra
 ining data or specially crafted inputs. \n\nTo accomplish this, the postdo
 c will use tools from AI to develop characterizations of dynamic systems, 
 will develop the underlying math for adversarial manipulation of the ML mo
 dels, and methods that will seek to enable detection and prevention of man
 ipulation via attacking the learning methods. In the context of the power 
 grid, the postdoc will examine ways in which automated power grid systems 
 can be attacked, and will develop algorithms to determine which cyber-phys
 ical system properties should be adjusted to re-stabilize unstable operati
 ng points.  \n\nImportant qualities of the position include experience in 
 adversarial learning and deep learning, experience with control theory, an
 d experience with software development, including experience in working in
  team-based development environments; and a strong interest in science, en
 abling scientific research, learning new scientific domains, and working c
 losely with domain scientists.\n\nThe Data Science and Technology (https:/
 /crd.lbl.gov/departments/data-science-and-technology/) department at Berke
 ley Lab develops software and tools to enable scientists to address comple
 x and large-scale computing and data analysis problems beyond what is poss
 ible today. DST engages in partnerships with scientists to understand thei
 r computing and data analysis challenges to develop leading-edge solutions
 . Our research areas address aspects of scientific computing that are not 
 adequately addressed by existing frameworks and tools. Details on current 
 and recent projects are available on http://dst.lbl.gov and http://dst.lbl
 .gov/security.\n\nWhat You Will Do:\n• Write scientific research papers su
 itable for submission to peer-reviewed computer science venues, such as IC
 ML, ICLR, or NeurIPS; computer security venues such as the IEEE Symposium 
 on Security & Privacy or the USENIX Security Symposium; or control theory 
 venues.\n• Use tools from AI to develop reduced order characterizations of
  high-dimensional dynamic systems.\n• Develop the math underlying adversar
 ial manipulation of the ML models and methods that will seek to enable det
 ection and prevention of manipulation via attacking the learning methods. 
 \n• In the context of the power grid, examine ways in which automated powe
 r grid systems can be attacked, and will develop algorithms to determine w
 hich cyber-physical system properties should be adjusted to re-stabilize u
 nstable operating points. \n• Work with research staff in the Berkeley Lab
  Data Science & Technology Department and Grid Integration Group, with res
 earchers and application scientists throughout the Berkeley Lab and the DO
 E Office of Science community, and with faculty and student collaborators 
 from universities throughout the world.\n\nWhat is Required:\n• Ph.D. degr
 ee in Computer Science, Computer Engineering, Electrical Engineering, Math
 ematics, Statistics, or a related technical field is required.\n• An estab
 lished track record of peer-reviewed publications in deep learning and/or 
 adversarial machine learning.\n• Experience with key tools used in scienti
 fic data discovery, such as Jupyter notebooks, Spark, PyML, TensorFlow, an
 d/or related software systems.\n• Proven experience writing software and p
 roficiency and experience in programming languages such as C/C++ and/or Py
 thon.\n• Demonstrated ability to work independently and collaboratively in
  a diverse interdisciplinary team and contribute to an active intellectual
  environment.\n• Excellent written and verbal skills.\n• Keen interest in 
 solving science challenges.\n\nDesired Qualifications:\n• Experience in co
 ntrol theory and/or signal processing.\n• Experience with electric power s
 ystems is a plus.\n• Proficiency with UNIX tools and computer systems.\n\n
 Notes:\n• This is a full-time 2 year, postdoctoral appointment with the po
 ssibility of renewal based upon satisfactory job performance, continuing a
 vailability of funds and ongoing operational needs. You must have less tha
 n 3 years of paid postdoctoral experience. Salary for Postdoctoral positio
 ns depends on years of experience post-degree.\n• This position is represe
 nted by a union for collective bargaining purposes.\n• Salary will be pred
 etermined based on postdoctoral step rates.\n• This position may be subjec
 t to a background check. Any convictions will be evaluated to determine if
  they directly relate to the responsibilities and requirements of the posi
 tion. Having a conviction history will not automatically disqualify an app
 licant from being considered for employment.\n• Work will be primarily per
 formed at Lawrence Berkeley National Lab, 1 Cyclotron Road, Berkeley, CA.\
 n\nHow To Apply\nApply directly online at http://50.73.55.13/counter.php?i
 d=189792 and follow the on-line instructions to complete the application p
 rocess.\n\nEqual Employment Opportunity: Berkeley Lab is an Equal Opportun
 ity/Affirmative Action Employer. All qualified applicants will receive con
 sideration for employment without regard to race, color, religion, sex, se
 xual orientation, gender identity, national origin, disability, age, or pr
 otected veteran status. Berkeley Lab is in compliance with the Pay Transpa
 rency Nondiscrimination Provision under 41 CFR 60-1.4.  Click here (https:
 //www.dol.gov/agencies/ofccp/posters) to view the poster and supplement: "
 Equal Employment Opportunity is the Law."\n\nLawrence Berkeley National La
 boratory encourages applications from women, minorities, veterans, and oth
 er underrepresented groups presently considering scientific research caree
 rs.\n\nRegistration Category: Tech Program Reg Pass, Workshop Reg Pass, Tu
 torial Reg Pass, Exhibits Reg Pass
END:VEVENT
END:VCALENDAR