-
Notifications
You must be signed in to change notification settings - Fork 0
/
cv.html
125 lines (89 loc) · 7.14 KB
/
cv.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
<!DOCTYPE html>
<html lang="en">
<head>
<title>Henrik Siljebråt - cv -</title>
<meta charset="utf-8" />
<meta name="viewport" content="width=device-width, initial-scale=1" />
<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
<meta name="description" content="Henrik Siljebråt has a PhD in Cognitive Neuroscience from Goldsmiths, University of London. Henrik likes animal and artificial intelligence, and can help you spin the stories you want the world to remember." />
<meta name="keywords" content="Henrik Siljebråt, Henrik Siljebrat, artificial intelligence, cognitive science, learning, decision-making, python, cognition, intelligence, bayesian inference, writer, talespinner, phd, computational neuroscience, ai consultant">
<link rel="stylesheet" href="css/stylish.css" />
<link rel="stylesheet" href="css/stylish-page.css" />
<link rel="canonical" href="https://henrik.siljebrat.se/cv.html" />
<link rel="icon" type="image/png" href="images/logo.png">
</head>
<body>
<div class="mainer-container">
<header class="header-container">
<div class="header-title">
<a href="index.html"><img src="avatars/threeye.jpg" alt="logo a.k.a. my face" class="header-logo"></a>
<a href="index.html"><p class="header-name">Henrik Siljebråt</p></a>
</div>
<nav>
<ul class="navigation-menu">
<li><a href="index.html">Home</a></li>
<li><a href="https://gamescapad.es">Blog</a></li>
<li><a href="cv.html">CV</a></li>
<li><a href="projects.html">Projects</a></li>
<li><a href="about.html">About</a></li>
</ul>
</nav>
</header>
<article>
<h1>CV<br />or: How I learned to stop worrying and love the AI</h1>
<section class="main-text">
<p>You can download a pdf of my CV by clicking <a href="files/siljebrat_cv.pdf">here</a></p>
Timeline:
<ul>
<li><a href="https://www.nordaxon.com">AI Consultant</a></li>
<li><a href="#phd">PhD in Computational Neuroscience</a></li>
<li><a href="#masters">Master in Cognitive Science</a></li>
</ul>
</section>
<h2 id="phd">PhD in Computational Neuroscience</h2>
<section class="main-text">
<p>I got my PhD from <a href="https://gold.ac.uk">Goldsmiths, University of London</a>. My thesis is called:
<blockquote class="thesis-title">On State Representations and Behavioural Modelling Methods in Reinforcement Learning</blockquote>
If this sounds fun, you can <a href="files/henrik_siljebrat_thesis.pdf">download my thesis</a>. There are many pretty graphs!</p>
<h3>What does it mean to have a PhD?</h3>
<p>It's common that people outside academia are not aware of the diverse set of skills you aquire during a PhD. Some examples of such skills that I now have:
<ul>
<li>Learn new things quickly</li>
<li>Scope projects to get them done</li>
<li>Adapt information for the intended audience</li>
<li>Write; both technically and creatively</li>
<li>Teach</li>
<li>Present/Speak in public</li>
<li>Communicate across fields/departments</li>
</ul>
</p>
<h3>My research in a nutshell</h3>
<p>To build a robot that learns and makes decisions, what algorithms would we need?</p>
<h3>My research in a coconut</h3>
<p>Reinforcement learning (RL) - algorithms for learning from rewards - has proved successful in the cognitive sciences for explaining both neuronal signals and behaviour in animals, and for producing impressive results in artificial intelligence.</p>
<p>Essential to RL models are state representations. Based on what current state an animal or artificial agent is in, they learn optimal actions by maximizing future expected reward. But how are humans able to learn and create representations of states?</p>
<p>This thesis approaches this quesetion from two fronts. First, we thoroughly investigate methods for fitting behavioural models to human lab data. In contrast to recent proposals, we find that the best methods for model selection - determining what model most likely generated some data - are based on maximum likelihood estimation, rather than Bayesian inference. We also demonstrate the importance of considering individual differences in model fitting: the model which best fits the performance of one participant may not fit the behaviour of another participant.</p>
<p>Second, we introduce Shapetask - a novel learning and decision-making task where participants must find hidden structure in a sequence, without the task explicitly rewarding the appropriate actions. We show that some humans can find this pattern, while RL cannot, unless equipped with appropriate state representations. We then show how previously proposed models that integrate RL with complex state representations can account for individual human behaviour in the Shapetask.</p>
<p>We argue our results add to the growing literature indicating a broader role for dopamine as one involving general sensory prediction errors, not just reward prediction errors. Further, we argue Shapetask holds promise for use in further research on the topic of state representation and task structure. Such research may illuminate the workings of animal brains, and contribute to artificial intelligence, where enhanced models of state representations could improve data efficiency and generalisability over current generation systems.</p>
<h3>My research in papers</h3>
<ul>
<li><a href="files/siljebrat_pickering-2020-state_representations_shape_sequence_task.pdf">The Effect of State Representations in Sequential Sensory Prediction: Introducing the Shape Sequence Task</a><br />(2020, CogSci Conference, Peer-reviewed)</li>
<li><a href="files/siljebrat_et_al-2018-towards_human_like_artificial_intelligence.pdf">Towards human-like artificial intelligence using StarCraft II</a><br />(2018, FDG Conference, Peer-reviewed)</li>
</ul>
<p>My PhD studies were fully funded by the <a href="https://epsrc.ukri.org">EPSRC</a> via <a href="https://iggi.org.uk">IGGI</a>.</p>
<h2 id="masters">I'm a Cognitive Scientist</h2>
<section class="main-text">
<p>I was trained (potentially brainwashed) by the brilliant people at <a href="https://www.lucs.lu.se/">Lund University Cognitive Science</a> to see connections between all forms of thinking systems and how they interact.</p>
<div class="cogsci-container">
<img src="images/cogsci.png" alt="cognitive science - linguistics, neuroscience, artificial intelligence, philosophy, anthropology, psychology and ethology" width="250" height="250">
<p>This image borrowed from Wikipedia is the best I've found that shows how cognitive science is a mix of many fields. However, it left out animal behaviour and biology; ethology. So I added a bumblebee. Obviously a spider would have made more sense sitting in this web. But bumblebees are cuter.</p>
</div>
</section>
<!-- <h1>I'm a Creator</h1>
<section class="main-text">
<p>I write, program and sometimes draw to make stuff up.</p>
</section> -->
</article>
</div>
</body>
</html>